REVIEWS

Deubiquitylating and drug discovery: emerging opportunities

Jeanine A. Harrigan1*, Xavier Jacq1*, Niall M. Martin1,2, and Stephen P. Jackson1,3 Abstract | More than a decade after a Nobel Prize was awarded for the discovery of the system and clinical approval of proteasome and ubiquitin E3 inhibitors, first-generation deubiquitylating (DUB) inhibitors are now approaching clinical trials. However, although our knowledge of the physiological and pathophysiological roles of DUBs has evolved tremendously, the clinical development of selective DUB inhibitors has been challenging. In this Review, we discuss these issues and highlight recent advances in our understanding of DUB enzymology and biology as well as technological improvements that have contributed to the current interest in DUBs as therapeutic targets in diseases ranging from oncology to neurodegeneration.

10 Ubiquitin The sequential enzymatic processes that covalently exist in eukaryotic cells . The recent demonstration of A small that is attach ubiquitin, a 76‑residue polypeptide, to target post-translational modification of ubiquitin itself pro‑ conjugated to other proteins — a process known as ubiquitylation — are vides an additional layer of regulation that affects various (including itself) as a now well understood1 (FIG. 1a). In some cases, a single cellular processes11. post-translational modification, ubiquitin is attached to the target protein, whereas in Like other post-translational modifications, ubiquity­ often to control cellular signalling or degradation of the others, multiple monoubiquitin adducts are conjugated lation is reversible: peptidases termed deubiquitylating modified protein. to different residues of the target. In many instances, enzymes (DUBs) can cleave ubiquitin from substrate pro‑ various types of ubiquitin chains are produced, wherein teins, edit ubiquitin chains and process ubiquitin precur‑ Ubiquitin-like proteins one ubiquitin moiety is attached to a free amino group sors12. Some DUBs and related enzymes are involved in (UBLs). Proteins such as small 13 ubiquitin-like modifier (SUMO) of another. This leads to linear ubiquitin chains and editing or processing UBLs and their conjugates ; prime or interferon-stimulated chains involving internal ubiquitin lysine residues K6, examples of these being the SENP (sentrin/SUMO- 15 (ISG15) that adopt K11, K27, K29, K33, K48, K63, as well as mixed ubiq‑ ­specific protease) proteins that process SUMO precursors a β‑grasp fold, which is uitin chains containing different linkages, or linkages and SUMO conjugates14. DUBs are classified into six fam‑ characteristic of ubiquitin and between ubiquitin and ubiquitin-like proteins (UBLs) ilies based on sequence and domain conservation (FIG. 1b): related proteins. that include small ubiquitin-like modifier (SUMO) USPs (ubiquitin-specific proteases), UCHs (ubiquitin and neuronal precursor cell-expressed developmentally carboxy-terminal ), MJDs (Machado–Josephin downregulated protein 8 (NEDD8). domain-containing proteases), OTUs (ovarian tumour 1Mission Therapeutics Ltd, Moneta, Babraham Research These different types of ubiquitin and UBL modi‑ proteases), MINDYs (motif-­interacting with ubiquitin-­ Campus, Cambridge CB22 fications, sometimes referred to as ‘the ubiquitin code’, containing novel DUB family) and JAMMs (JAB1, MPN, 3AT, UK have specific and diverse effects on protein and cell MOV34 family). SENPs and the first five DUB fami‑ 2Present address: Artios physiology. For example, such modifications can target lies are cysteine peptidases, whereas JAMMs are zinc Pharmaceuticals Ltd, Maia, proteins that are damaged or improperly folded, or that metallopeptidases. Babraham Research Campus, Cambridge CB22 3AT, UK have intrinsically short half-lives for degradation via the Ubiquitylation and related processes control myr‑ 3The Wellcome Trust and ubiquitin–proteasome system (UPS)2. Appropriately iad aspects of human cell biology and physiology, and Research UK Gurdon polyubiquitylated proteins are recognized and degraded defects in such processes contribute to many diseases. Institute, and Department of by the 26S macromolecular proteasome complex3 via Accordingly, DUB deregulation contributes to vari‑ Biochemistry, Tennis Court Road, University of Cambridge, mechanisms that have been extensively reviewed else‑ ous sporadic and genetic disorders. Notable examples 4,5 Cambridge CB2 1QN, UK. where . In other instances, ubiquitylation regulates include: the UCH family member BRCA1‑associated *These authors contributed protein interactions, localization and enzymatic activ‑ protein 1 (BAP1), mutated in melanoma, meso­thelioma equally to this work. ities, thereby affecting cellular processes, including and renal cell carcinoma15; USP6, translocated in aneu‑ 16 Correspondence to S.P.J. transcription, DNA damage signalling and DNA repair, rysmal bone cysts ; USP7, mutated in neurological [email protected] progression, endocytosis, and vari‑ disorders17; USP8, whose cause Cushing 6–9 18,19 doi:10.1038/nrd.2017.152 ous other processes . Such control mechanisms often disease ; USP9X, whose mutations cause develop‑ Published online 29 Sep 2017 involve ubiquitin-binding proteins, many of which mental disorders20 and whose expression is dysregulated

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 57 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS a in cancer21; USP15, amplified in certain glioblastoma, Ub Ub Ub Ub Ub Ub Ub Ub Ub breast and ovarian cancers22; and CYLD, commonly Ub Ub Ub Ubiquitin mutated in cylindromatosis23. Deregulation of MJD- Ub L precursors family DUBs has also been linked to diseases associated Ub S with polyglutamine amplification. For example, expan‑ sion of DNA ‘CAG’ trinucleotide repeats in (ATXN3) causes Machado–Joseph disease (also known DUB as 3)24. Furthermore, mutations in the JAMM family member associated molecule with the SH3 domain of STAM (AMSH; also known as STAMBP) Ub Ub E1 25 C cause microcephaly–capillary malformation syndrome . Ub Ub There has been growing interest in exploiting com‑ DUB ATP ADP + P i ponents of the ubiquitylation machinery as therapeutic targets26. Although there has been strong progress in developing small-molecule inhibitors of ubiquitin and Ub Ub 27 Ub Ub Ub UBL E1 enzymes , the highly pleiotropic nature of E1s Substrate means that such drugs will likely be confined to acute K E2 K settings, such as in the treatment of aggressive . C Substrate Ub Given their greater numbers and diversity, E2s, E3s and DUBs offer the potential for developing drugs with more specific effects. In particular, being a group of diverse enzymes with well-defined catalytic clefts, DUBs are intrinsically attractive as potential drug targets26. E3 E3 E2 E2 However — as we discuss further below — until recently, C C the development of selective DUB inhibitors has been Ub Ub

Substrate Figure 1 | The ubiquitylation cascade and the deubiquitylase family of proteins. a | Schematic of key events in ubiquitylation and deubiquitylation. The E1 enzyme activates ubiquitin in an ATP-dependent manner, resulting in a covalent thioester linkage between ubiquitin and the E1 cysteine residue. Ubiquitin is then b USP37 USP20 USP33

USP48 USP28 USP25 transferred to an E2 conjugating enzyme, forming a USP9Y USP9X USP24 thioester linkage with the catalytic cysteine. Finally, an E3

USP26 USP29 USP5 USP34 USP13 USP39 USP47 USP44 ligase assists or directly catalyses the transfer of ubiquitin USP49 USP36 USP7 USP42 USP40 USP17-like from the E2 to a substrate, usually via a lysine side chain. CYLD DUB3 USP52 USP30 An example of a HECT (homologous to the E6AP carboxyl USP10 USP53 USP18 terminus) or RBR (RING-between-RING) E3 ligase is shown. USP54 USP41 USP15 USP8 In subsequent rounds, ubiquitin molecules can be USP21 USP4 USP14 conjugated to the N‑terminal amino group or lysines on USP11 USP50 ubiquitin itself to form chains. Deubiquitylating enzymes USP19 USP51 USP43 USP22 (DUBs) remove ubiquitin molecules from substrates or USP31 USP27X process ubiquitin precursors to generate free ubiquitin USP45 USP3 USP2 pools. b | DUB phylogenetic tree. Sequences for full-length USP16 USP12 DUB and SENP (sentrin/SUMO-specific protease) proteins USP1 USP46 USP38 USPL1 were aligned with COBALT (constraint-based multiple USP35 DESI1 alignment tool), a computational tool for multiple protein USP32 DESI2 sequences, and subsequently visualized with FigTree v1.4.3. USP6 SENP6 SENP7 In regard to USP17-like, note that various related human UCHL3 SENP3 USP17-like DUBs exist. AMSH, associated molecule with UCHL1 SENP5 UCHL5 SENP8 the SH3 domain of STAM; AMSHLP, AMSH-like protease; BAP1 SENP1 ATXN3, ataxin 3; BAP1, BRCA1‑associated protein 1; SENP2 MINDY2 MYSM1 CEZANNE, cellular zinc finger anti-NF-κB protein; CSN, MINDY1 MPND PRPF8 COP9 signalosome complex subunit; CYLD, MINDY4 AMSH MINDY3 AMSHLP cylindromatosis; DESI, desumoylating isopeptidase; EIF3, JOSD2 BRCC3 JOSD1 CSN5 PSMD14 eukaryotic translation initiation factor 3; JAMM, JAB1, MPN, EIF3S3 ATXN3L ATXN3 CSN6EIF3S5 MOV34 family; JOSD, josephin domain; MINDY, PSMD7 OTUB1 OTUB2

A20 OTULIN FAM105A motif-interacting with ubiquitin-containing novel DUB OTUD6B YOD1 OTUD6A OTUD1 family; MJD, Machado–Josephin domain-containing OTUD5

OTUD3 VCPIP1 OTUD4 protease; OTUD, OTU domain-containing protein; PRPF8,

TRABID

CEZANNE pre-mRNA-processing splicing factor 8; UCHL, ubiquitin CEZANNE2 carboxy-terminal -like; USP, ubiquitin-specific USP SENP JAMM OTU MJD MINDY UCH protease; VCPIP1, valosin-containing protein p97/p47 complex-interacting protein 1.

Nature Reviews | Drug Discovery 58 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

limited by insufficient understanding of DUB biology, carcinomas and correlates with overexpression and difficulties in establishing robust biochemical assays suit‑ tumour growth41. PSMD14 has also been reported to able for compound screening, limitations in cellular and regulate the ubiquitylation and stability of the in vivo models to assess DUB activity or inhibition, and the receptor tyrosine kinase ERBB2 (REF. 42). Furthermore, pleiotropic nature of various small-molecule DUB as PSMD14 has been connected to promoting cellular inhibitors. With many of these issues now being largely responses to DNA double-strand breaks, particularly overcome, the rate of progress in DUB drug discovery has by the process of , PSMD14 accelerated over the past few years, with various selective inhibition could potentially sensitize cancer cells to compounds being described and characterized by both DNA-damaging agents and/or preferentially kill cancer academic groups and companies. cells that rely strongly on homologous recombination43. In this Review, we discuss how DUBs and their Another potential anticancer therapeutic target is deregulation affect human diseases, particularly can‑ USP14, which is primarily associated with the protea‑ cer, neurodegeneration and inflammation (TABLE 1), and some 19S regulatory subunit, where it potentiates ubiq‑ highlight the therapeutic potential for pharmacological uitin recycling44. USP14 is not constitutively active but modulation of DUB activities. Recent advances in assay reversibly associates with the 19S RPN1 subunit (also development and screening technologies, which are known as PSMD2), which enhances its activity45. USP14 enabling researchers and drug developers to overcome inhibits proteasomal degradation of ubiquitin–protein recurrent challenges in the clinical translation of DUB conjugates by trimming ubiquitin chains on protein sub‑ inhibitors, are also discussed. strates before their degradation46. USP14 expression is upregulated in non-small-cell , especially in DUBs in oncology adenocarcinoma47, and its levels are reportedly elevated Accumulating evidence implicates DUBs in tumori­ in ovarian cancer samples48. In line with this, USP14 genesis at multiple levels (FIG. 2). First, DUBs such as is connected to several important signalling pathways, BAP1, UCHL1 and CYLD have been described as dis‑ for example, as a substrate of AKT mediating intra­ playing intrinsic oncogenic or tumour suppressor activi‑ cellular signalling for growth factors49 and a modula‑ ties28. Second, some DUBs, such as USP22, are connected tor of Dishevelled proteins, key positive regulators of to controlling key epigenetic changes that promote WNT signalling50. tumour development29. Third, through their deubi­ Like USP14, the DUB UCHL5 reversibly interacts quitylating activities, various DUBs, such as USP7 and with the proteasome51, binding to the RPN13 receptor USP28, have been reported to regulate the levels and/or (also known as ADRM1)52 in a manner that enhances Proteasome activities of various oncogene or tumour suppressor pro‑ UCHL5 isopeptidase activity51,53. A key function A protein complex that 30,31 recognizes and degrades teins . Fourth, DUBs modulate other therapeutically of UCHL5 is to remove distal ubiquitin moieties from polyubiquitylated proteins. relevant cellular components and processes, such as the polyubiquitylated proteins, thereby liberating proteins The 19S regulatory subunit UPS (for example, USP14 and UCHL5 (also known from destruction54, or facilitating destruction of cer‑ contains ATPase- and as UCH37))32, stem cell renewal (for example, USP16 tain substrates, as described for inducible nitric oxide ubiquitin-binding sites, 29,33 55 and the 20S catalytic core or USP22) , DNA damage response (DDR) and synthase and nuclear factor-κB inhibitor-α (IκBα) . It 9 contains proteases. DNA repair (for example, USP1 or USP11) , immuno-­ therefore seems that, like USP14, UCHL5 suppresses oncology (for example, USP7)34 or receptor tyrosine the destruction of certain proteins, while promoting the Deubiquitylating enzymes kinases (for example, USP8 or USP9X)35,36. Consequently, degradation of others. Notably, RNA interference stud‑ (DUBs). Proteins that remove and as described in more detail below, various DUBs are ies showed that depletion of either USP14 or UCHL5 ubiquitin from substrate proteins or cleave ubiquitin emerging as attractive targets for the development of alone had no detectable effect on , proteas‑ precursors. novel cancer therapies. ome structure or proteolytic capacity but did accelerate cellular protein degradation53. By contrast, depletion of Tumour suppressor Proteasomal DUBs both DUBs decreased protein degradation, suggesting A type of protein that normally restricts the proliferation or The successful targeting of the proteasome for cancer that they have overlapping functions. UCHL5 is over­ invasive properties of normal therapy is underlined by the clinical success of borte‑ expressed in epithelial ovarian cancer, which is asso‑ cells. Mutations or zomib, a broadly acting proteasome inhibitor, in refrac‑ ciated with advanced tumour progression and poor loss-of-expression of tumour tory multiple myeloma37 or mantle cell myeloma38. clinical outcome56. UCHL5 is also overexpressed in suppressors can lead to However, three DUBs associated with proteasome func‑ hepatocellular carcinoma, and was shown to promote cancer progression. tions — PSMD14 (also known as POH1), USP14 and cell migration and invasion57. Oncogene UCHL5 — may represent more specific anticancer tar‑ These proteasome-associated DUBs represent A type of protein that normally gets. To facilitate the degradation of proteasome-­targeted attractive drug targets, as their inhibition might have controls growth, differentiation substrates, these specialized DUBs remove ubiquitin substantial effects on cancer cell physiology but with or survival of cells but whose or expression at moieties that would otherwise impede entry into the fewer toxicities than are seen with drugs targeting core 39 58 abnormally high levels 20S proteasome catalytic core . proteasome catalytic function . Indeed, VLX1570 promotes tumorigenesis. The JAMM metalloprotease PSMD14 has been high‑ (TABLE 2), the most advanced reported DUB inhibitor, lighted as a potential therapeutic target through studies which was recently in phase I trials (now suspended) Isopeptidase showing that its levels inversely correlate with survival for treatment of multiple myeloma and solid tumours59, A type of enzyme that (REF. 60) hydrolyses amide bonds that of patients with multiple myeloma and that its deple‑ has been described to target USP14 and UCHL5 . 40 occur outside of the main chain tion impairs proliferation of multiple myeloma cells . In VLX1570 is a ring-expanded version of the compound in a polypeptide chain. addition, nuclear PSMD14 is elevated in hepatocellular b-AP15 (also known as VLX1500) that was identified

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 59 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Table 1 | DUBs connected with human diseases Process Targeted DUB Target Rationale Disease association Refs Oncology Proteasome PSMD14 Many General protein turnover Liver cancer 41 USP14 Lung and ovarian cancers 47,48,240 UCHL5 Oesophageal and ovarian cancers 56,240,241 DNA repair USP1 FANCD2, PCNA • Fanconi anaemia pathway Osteosarcoma 73 • Translesion synthesis USP4 CTIP Homologous recombination Lung, breast and liver cancers 242–244 USP11 PALB2 Breast cancers 245 USP9X Claspin Replication checkpoint Sarcoma and tumours of the colon, cervix, 21 kidney, breast, prostate and brain and ATXN3 , HDM2 Promotes p53‑mediated p53‑expressing tumours 102 tumour suppressors apoptosis CYLD NF‑κB Unclear • Mutated in cylindromatosis and multiple 23, myeloma 280–283 • Reduced expression in colon and liver cancers and melanoma UCHL1 AKT Osteosarcoma, myeloma and tumours of 252–258 the colon, breast, lung and kidney USP6 – Translocated in aneurysmal bone cysts 284 USP7 p53, HDM2 HDM2‑overexpressing tumours Leukaemia and ovarian and lung cancers 246–249 USP8 EGFR Regulates recycling of • Lung cancer 18,19, receptor tyrosine kinases, • Mutated in Cushing syndrome 107,285 including EGFR USP15 Type I TGFβ Regulation of TGFβ signalling Glioblastoma and breast and ovarian cancers 22,112 receptor, R‑SMADs USP20 HIF1α Sensitizes hypoxic tumour cells – 290 USP28 FBW7, , JUN, APC-driven cancers Colorectal and ovarian cancers 250,251 Notch Epigenetics BAP1 , Epigenetic deregulation of Uveal melanoma, sporadic melanoma, 286–289 HCF1 tumours mesothelioma and kidney cancer USP22 Histone H2A Colorectal, breast, oesophageal, lung and 127–130, pancreatic cancers 259,260 CNS disorders Neurodegeneration ATXN3 Counteracts Parkin Expansion of CAG trinucleotide repeats 151 autoubiquitylation causes Machado–Joseph disease USP7 α‑Synuclein, REST • Antagonizes ubiquitylation of – 149 α‑synuclein • Regulates REST signalling and neuronal differentiation USP8 Parkin, K6‑linked • Regulates mitophagy by – 148, Ubiquitin chains removing ubiquitin from 263–265 Parkin • Regulates TRKA levels in an NGF-dependent manner USP14 Proteasome Increased clearance of proteins Mutations cause ataxia 44,262 substrates involved in neurodegeneration (Tau or ATXN3) USP15 – Opposes Parkin-mediated Glioblastoma 143,150 mitophagy USP30 Ubiquitin Mitochondrial dysfunction, – 142,144, conjugates at mitophagy 146,261 mitochondrial surface, Parkin Down syndrome USP16 Histone H2A Antagonizes self-renewal – 158–160, and/or senescence in Down 266 syndrome

60 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Table 1 (cont.) | DUBs connected with human diseases Process Targeted DUB Target Rationale Disease association Refs Inflammation, immunity and infectious disease Negative regulation A20 NEMO, RIPK1, Inhibits NF‑κB signalling Expression levels regulated by TNFα, IL‑1β 167, of the immune TRAF6 and LPS 273,274 response CYLD RIG1, TBK1, IKKε – 170,171 OTULIN RIPK1, RIPK2, – 172,173 NEMO USP18 – • Functions in haematopoietic Expression regulated by IFNγ 268, cell differentiation 269,291 • Removes ISG15 conjugates • Negative feedback regulator of type I IFN signalling USP25 RIG1, TRAF2, • Negatively regulates Expression regulated by IFN and IRF7 270–272, TRAF3, TRAF6 IL‑17‑triggered signalling 292,303 • Negatively regulates virus-induced type I IFN production • Positive feedback regulation of innate immune responses against RNA and DNA viruses

Treg responses USP7 FOXP3 • Stabilizes FOXP3 in Expressed and regulated upon viral 34,267,293 Treg cells infections in B and T cells • Negative regulator of TNFα-stimulated NF‑κB activity

USP21 FOXP3 Stabilizes FOXP3 in Treg cells – 131

TH1 and TH17 CEZANNE ZAP70 • Positive regulator of T cell – 193 responses receptor signalling • Binds to and deubiquitylates ZAP70 TRABID JMJD2D Positive regulator of IL‑22 and – 191 IL‑23 production

+ USP4 RORγt, RIG1, • Stabilizes RORγt in TH17 cells Highly expressed in CD4 T cells from 189, TAK1 • Positively regulates patients with rheumatic heart disease 275,276, RIG1‑mediated antiviral 294 response • Negative regulator of TLR–IL‑1R signalling • Targets TAK1 to downregulate TNFα-induced NF-κB activation

USP10 T‑bet Stabilizes T‑bet in TH1 cells Highly expressed in PBMCs from patients 196 with asthma USP17 RORγt, RIG1, • Positive regulator of RORγt in – 277–279, IL‑33 TH17 cells 295 • Regulates virus-induced type I IFN signalling • Regulates the stability and nuclear function of IL‑33 USP18 TAK1–TAB1 Regulates TAK1–TAB1 Expression induced by 192

complex interaction required for TH17 differentiation

APC, adenomatous polyposis coli protein; ATXN3, ataxin 3; BAP1, BRCA1‑associated protein 1; CEZANNE, cellular zinc finger anti-NF-κB protein; CNS, central nervous system; CYLD, cylindromatosis; CTIP, C-terminal-binding protein-interacting protein; EGFR, epidermal growth factor receptor; FANCD2, Fanconi anaemia group D2 protein; FBW7, F-box and WD40 domain-containing protein 7; FOXP3, forkhead box protein P3; HCF1, host cell factor 1; HDM2, human double minute 2; HIF1α, hypoxia-inducible factor 1α; IKKε, IκB kinase-ε; IFN, interferon, IL, interleukin; IRF7, interferon regulatory factor 7; ISG15, IFN-stimulated gene 15; JMJD2D, Jumonji domain-containing protein 2D; LPS, lipopolysaccharide; NEMO, NF-κB essential modulator; NF-κB, nuclear factor-κB; NGF, β-nerve growth factor; PALB2, partner and localizer of BRCA2; PBMC, peripheral blood mononuclear cell; PCNA, proliferating cell nuclear antigen; REST, RE1‑silencing ; RIG1, retinoic acid-inducible gene 1‑like receptor 1; RIPK, receptor-interacting serine/threonine-protein kinase; RORγt, retinoid-related orphan receptor-γt; R-SMAD, receptor-regulated SMAD; TAB, TAK1‑binding protein; TAK1, TGFβ-activated kinase 1; TBK1, TANK-binding kinase 1; TGFβ, transforming growth factor-β; TH cells, T helper cells; TLR, Toll-like receptor; TNFα, tumour necrosis factor-α; TRAF6, TNF receptor-associated factor 6; Treg cells, regulatory T cells; TRKA, tropomyosin-related kinase A; UCHL, ubiquitin carboxy-terminal hydrolase-like; USP, ubiquitin-specific protease; ZAP70, 70 kDa ζ-chain-associated protein.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 61 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Cytoplasm Proteasomal degradation GFR IU1

Ub p53 USP14 Proteasome Ub AKT PI3K USP7 HDM2 HDM2 VLX1570 USP7 UCHL1 LDN-57444 PSMD14 p53 UCHL5

ADC-01, ADC-03, HBX41108, HBX19818, P5091, P22077

DNA repair DNA repair Ub Ub Mitoxantrone FANCD2 PCNA Proteasomal degradation

USP1 USP11 Ub PALB2 PALB2 Ub FANCD2 PCNA MYC BRCA1 BRCA2 BRCA2 ML323

USP28

DNA damage

RAD50 DNA repair DNA MYC Vialinin A MRE11 NBS1 Ub CTlP Replication stress Ub USP4 USP22 Claspin Claspin Ub USP9X BAP1 Ub

Chromatin remodelling WP1130 Nucleus

Figure 2 | Various roles of DUBs in oncology. Selected, representative protein 1 (BAP1) and USP22 participate in chromatin remodelling by examples of deubiquitylating enzymes (DUBs; light blue ovals) involved in deubiquitylating histones, and ubiquitin carboxy-terminal hydrolase-like 1 distinct cellular pathways and regulation of various ubiquitylated (UCHL1) plays a part in AKT signalling. TheseNature are Reviews representative | Drug Discovery examples substrates (dark blue ovals) related to oncology. The proteasome and only and not meant to be exhaustive. Examples of small-molecule associated DUBs facilitate protein turnover and recycle ubiquitin. compounds targeting these DUBs are shown. BRCA1, breast cancer type Ubiquitin-specific protease 28 (USP28) regulates turnover of 1 susceptibility protein; CTIP, C-terminal-binding protein-interacting­ the oncogene product MYC, ataxin 3 (ATXN3) controls the stability of the protein; FANCD2, Fanconi anaemia group D2 protein; GFR, growth factor tumour suppressor p53, and USP7 regulates p53 and its E3 receptor; MRE11, meiotic recombination 11 homologue 1; NBS1, human double minute 2 (HDM2). USP1, USP4 and USP11 have important Nijmegen breakage syndrome protein 1; PALB2, partner and localizer of roles in DNA damage repair, whereas USP9X regulates claspin and is BRCA2; PCNA, proliferating cell nuclear antigen; PI3K, phosphoinositide linked to replication stress and checkpoint signalling. BRCA1‑associated 3‑kinase.

62 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

from cell-based screens looking for compounds induc‑ Another DUB linked to DNA repair is USP11, ing p53‑independent apoptosis. Cells treated with which was initially described to form a complex with b-AP15 accumulate polyubiquitin chains61, and it has the DDR tumour suppressor breast cancer type 2 been claimed that b-AP15 targets USP14 and possibly susceptibility protein (BRCA2) to promote the DNA also UCHL5 (REF. 60). This compound was reported double-strand break repair pathway of homologous to be reversible and reasonably selective against other recombination79. Depletion of USP11 has been shown DUBs60 in a cell-based activity probe assay, with a to sensitize cells to olaparib (also known as AZD2281),

median inhibitory concentration (IC50) of ~2 μM against which inhibits the DDR enzyme poly(ADP-ribose) purified 19S proteasome DUB activities. b-AP15 dis‑ polymerase 1 (PARP1)80. Recently, an interaction played strong activity when tested in various in vivo between BRCA1 and partner and localizer of BRCA2 solid tumour models59, including multiple myeloma62, (PALB2) — which functionally cooperates with but it remains to be seen whether VLX1570 selectivity BRCA2 in DNA repair — was shown to be under will be sufficient to deliver on its promise as a next-­ ubiquitin control, with PALB2 ubiquitylation sup‑ generation proteasome inhibitor. Cleave Biosciences has pressing its interaction with BRCA1 in a manner also published a series of patent applications describ‑ counteracted by USP11 (REF. 81). ing compounds that inhibit JAMM proteases, provid‑ The only currently reported USP11 inhibitor is ing potential angles for developing selective PSMD14 the topoisomerase inhibitor mitoxantrone82 (TABLE 2). inhibitors63–65 (TABLE 2). Although the authors reported low nanomolar potency in a pancreatic ductal cell survival DUBs linked to DNA repair model, no further development of this compound has One hallmark of cancer is the downregulation, loss or been reported. Given the apparent amenability of USP11 deregulation of certain DNA repair and DDR path‑ to small-molecule inhibition, it is notable that USP4, ways and/or strong reliance on such pathways66,67. DNA a DUB closely related to USP11, was recently shown to repair and DDR mechanisms are regulated by post-­ be involved in the DDR through promoting early stages translational modifications, such as ubiquitylation, and of homologous recombination83. many DUBs are strongly linked to such processes9,68. USP9X21, which maintains DNA replication fork sta‑ One example of this is USP1, a DUB identified bility and DNA damage checkpoint responses by regu‑ as a regulator of Fanconi anaemia group D2 protein lating the protein claspin during S phase84, may represent (FANCD2) ubiquitylation, a key protein involved in the another potential therapeutic target. USP9X has been Fanconi anaemia pathway of DNA crosslink repair69,70. shown to affect radiosensitivity in glioblastoma cells USP1 influences accumulation of the Fanconi anaemia by myeloid cell leukaemia 1 (MCL1)‑dependent and core complex at DNA damage sites and deubiquitylates -independent mechanisms85. The best-described USP9X FANCD2–FANCI in a cell cycle-dependent manner69. inhibitor is WP1130 (TABLE 2), identified in a screen for USP1 also removes monoubiquitin from proliferating Janus kinase 2 (JAK2) inhibitors, which was shown to cell nuclear antigen (PCNA), a DNA replication com‑ inhibit USP9X as well as other DUBs (USP5, USP14 ponent that also functions in DNA repair by translation and UCHL5)86,87. It was shown by mass spectrometry synthesis71. Other USP1 activities include involvement that the covalent mechanism of action of this compound in a feedback loop to limit DDR CHK1 protein kinase is reversible76. activity72, and the regulation of cellular differentiation in osteosarcoma cells by deubiquitylating and hence Regulation of oncogenes and tumour suppressors affecting the stability of inhibitors of DNA-binding Various DUBs have been reported to have connections proteins73. In vitro, USP1 activity is greatly stimulated to tumour-suppressing or oncogenic functions, and may by USP1‑associated factor 1 (UAF1; also known as therefore represent potential therapeutic targets88.

WDR48), enhancing USP1 catalytic turnover (kcat) but 74 not affinity (Km) for monoubiquitylated substrates . p53 regulation. Several DUBs have been linked to reg‑ Selective USP1 inhibitors with submicromolar potency ulation of the tumour suppressor protein p53, which have been identified75, and one of these, pimozide, was has pivotal roles in cellular stress responses and is lost shown to re‑sensitize platinum-resistant non-small-cell or mutated in many cancers89. Human double minute 2 lung cancer cells and promote FANCD2 and PCNA (HDM2; also known as ) is a RING-type ubiqui‑ monoubiquitylation75. However, although these studies tin E3 ligase and key negative regulator of p53 through indicated on‑target effects, DUB selectivity profiling sug‑ its ability to ubiquitylate p53 and target it for degrada‑ gested that pimozide might be less selective than initially tion90. By cleaving ubiquitin chains on HDM2, USP7 described76. Optimization of certain USP1 screening hits counteracts HDM2 proteasomal degradation, leading has generated additional molecules77, most notably a to p53 suppression through increased ubiquitylation selective pyrimidine core compound, ML323 (TABLE 2). and degradation91,92. In theory, USP7 inhibition should This molecule allosterically blocks complex formation therefore trigger HDM2 degradation, p53 stabiliza‑ between UAF1 and USP1 (REF. 78), potentiates cisplatin tion and ultimately activation of apoptotic pathways cytotoxicity, and increases PCNA and FANCD2 mono­ in tumour cells93. Additional USP7 targets have also ubiquitylation in cells77. So far, however, little progress been described, such as phosphatase and tensin homo‑ has been made in advancing selective USP1 inhibitors logue (PTEN), forkhead box protein O4 (FOXO4) into clinical development. and FOXP3 (REFS 34,94,95), suggesting alternative

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 63 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Table 2 | DUB inhibitors in development DUB Inhibitor Structure* Company/ Disease Stage of Refs Institution indication development PSMD14 8-Mercapto-N- Cleave Biosciences Oncology Preclinical 63–65 [(tetrahydro-3-furanyl) O methyl]-4-quinoline‑ carboxamide O NH

N SH

UCHL1 LDN‑57444 Brigham and Oncology Preclinical 120 O Women’s Hospital O and Harvard N Medical School Cl O N Cl

Cl

UCHL5 VLX1570 Vivolux Oncology Clinical trial 296 O2N O and NO2 phase (now USP14 suspended) F F N O

USP1 ML323 University of Oncology Preclinical 77,78,297 HN Delaware and N N N National Institutes N of Health N

USP2 ML364 O S National Institutes Inflammation Preclinical 301 of Health N N H

F3C NH O S O

USP4 Vialinin A HO OH Tokyo University Inflammation and Preclinical 189,190 of Agriculture and oncology HO OH Shanghai Institutes for Biological O O O O Sciences

64 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Table 2 (cont.) | DUB inhibitors in development DUB Inhibitor Structure* Company/ Disease Stage of Refs Institution indication development USP7 ADC‑01, ADC‑03 Unknown Almac Discovery Oncology, Preclinical 101 Immuno-oncology

HBX41108 (shown O N Hybrigenics Oncology, Preclinical 97,98 right), HBX19818 N Immuno-oncology

Cl N N

P5091 (shown right), O2N Progenra Oncology, Preclinical 298 P22077 Immuno-oncology

S S Cl O

Cl

USP8 9-(Ethoxyimino)-9H- Hybrigenics Oncology Preclinical 110,264 indeno[1,2-b]pyra‑ zine-2,3-dicarbonitrile O N N N

N N

USP9X WP1130 University of Oncology Preclinical 86 Michigan O

N H N N Br

USP10 Spautin 1 HN Shanghai Institute of Inflammation Preclinical 302 and Organic Chemistry F USP13 N F and Harvard Medical School N

USP11 Mitoxantrone H Thomas Jefferson Oncology Preclinical 82 N OH O HN OH University

OH O HN OH N H

USP14 IU1 and analogues O Harvard College Neurodegeneration Preclinical 46, N and Proteostasis 153–155, Therapeutics 299,300 F N

USP20 GSK2643943A N GSK Oncology Preclinical 106

NH F 2 N H

USP30 15‑oxospiramilactone O Chinese Academy of Neurodegeneration Preclinical 146 H Sciences

O O

OH H

*Chemical structures shown are representative only, and additional inhibitors and an assessment of their drug-likeness and reproducibility can be found in Kemp, 2016 (REF. 210). GSK, GlaxoSmithKline; UCHL, ubiquitin carboxy-terminal hydrolase-like; USP, ubiquitin-specific protease.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 65 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

therapeutic mechanisms for USP7 inhibitors. USP7 has USP20‑mediated deubiquitylation of HIF1α prevents also recently been shown to promote DNA replication proteasomal degradation, allowing for transcription via acting as a DUB for the UBL SUMO96. of hypoxic response . Thus, inhibition of USP20 The first published submicromolar USP7 inhibi‑ has potential for suppressing proliferation of hypoxic tor HBX41108 (REF. 97) was shown to be a rather non­ tumour cells. GlaxoSmithKline (GSK) presented brief specific inhibitor of DUBs76. Recently, more-selective details of its search for USP20 inhibitors at a conference amidotetrahydroacridine derivatives such as HBX19818 in 2012 (REF. 106) (TABLE 2). and HBX28258 were identified, although these exhib‑ ited fairly low potency98. Despite this, HBX19818 was EGFR and USP8. Ubiquitylation serves as a signal that shown to covalently bind to the catalytic Cys residue delivers membrane receptors from the cell surface to of USP7 in preference to other cysteinyl groups, and to lysosomes, and in mammalian cells, this mechanism stabilize p53 and promote G1 arrest and apoptosis in has been most intensively studied for epidermal growth cells98. Progenra’s thiophene chemical series also pro‑ factor receptor (EGFR). Upon EGF binding, activated vided relatively nonspecific USP7 inhibitors, including EGFR is rapidly internalized and transported, via the compounds P5091 and P22077 (REF. 99). In multi‑ early and late endosomes to lysosomes, where EGFR ple myeloma cells, P5091 stabilized p53 and inhibited is degraded. USP8 (also known as UBPY) deubiqui‑ tumour growth, whereas in animal models, P5091 was tylates EGFR on early endosomes, rescuing EGFR from well tolerated, inhibited tumour growth and prolonged degradation107,108. In several tumours — including glio‑ survival99. More recent in vivo studies using P22077 blastoma, lung and breast cancer — EGFR is amplified within an orthotopic neuroblastoma mouse model or mutated in the tyrosine kinase domain, resulting in showed statistically significant inhibition of xenograft deregulation of receptor signalling that drives uncon‑ growth100. While these findings are encouraging, lit‑ trolled proliferation of tumour cells109. USP8 inhibi‑ tle is known about the binding modes of these com‑ tors (for example, HBX90659) of a similar structural pounds and whether they can be further optimized into class to those identified for USP7 (REF. 110) have been more ‘drug-like’ entities. Recently, Almac Discovery reported (TABLE 2). Moreover, a derivative of these com‑ and Genentech reported that fragment-based screens pounds was shown to be efficacious in mouse models of provided hits as starting points for USP7 discovery pro‑ lung cancer111. grammes101. Optimization of one hit, ADC‑01, assisted by X‑ray crystallography, produced the non-covalent, TGFβ and USP15. USP15 regulates the transforming highly selective USP7 inhibitor ADC‑03 (TABLE 2). growth factor-β (TGFβ) pathway and is thought to be The stability of p53 has also recently been reported important for the proliferation of glioblastoma cells22. to be regulated by the DUB ATXN3 (REF. 102). ATXN3 USP15 binds to the SMAD7–SMAD-specific E3 ubi­ was shown to bind to and deubiquitylate p53, resulting quitin protein ligase 2 (SMURF2) complex, and deu‑ in p53 stabilization. Deletion of ATXN3 resulted in dest‑ biquitylates and stabilizes the type I TGFβ receptor, abilization of p53, whereas ectopic expression of ATXN3 leading to enhanced TGFβ signalling. The USP15 gene induced expression of p53 target genes and promoted is amplified in glioblastoma, breast and ovarian can‑ p53‑dependent apoptosis. How and whether ATXN3 cers, and high expression of USP15 correlates with high inhibitors could be exploited to treat cancer or other TGFβ activity22. Depletion of USP15 reduces the onco‑ diseases remains to be established. genic capacity of patient-derived glioma-initiating cells USP28 is another DUB that has recently been owing to diminished TGFβ signalling, suggesting ther‑ connected to p53, which functions together with apeutic potential for development of USP15 inhibitors. TP53‑binding protein 1 (TP53BP1) to promote p53‑­ In addition, USP15 has been shown to deubiquitylate mediated transcriptional responses103. Furthermore, receptor-regulated SMADs (R‑SMADs)112, another set USP28 is mutated in human cancer cells and is reported of TGFβ signalling pathway components. to antagonize the tumour suppressor F-box and WD40 domain-containing protein 7 (FBW7)31, highlighting UCHL1. The DUB UCHL1, normally expressed the potential for USP28 inhibitors in various tumours, predominantly in neurons and neuroendocrine tis‑ especially colorectal cancer104. USP28 has also been sues113,114, is highly expressed in many cancers, and its reported to antagonize ubiquitin-dependent degrada‑ expression correlates with poor prognosis115. Although tion of the oncogene product MYC as well as JUN and there are reports that UCHL1 has a tumour-suppressive Notch105. Although no USP28 inhibitors have yet been role, most evidence supports its role as an oncogene115. reported, it seems likely that drug discovery activities Indeed, in a transgenic mouse model with constitu‑ are underway. tively activated UCHL1, sporadic tumours developed in many tissues116. Moreover, in vitro tumorigenesis HIF1α and USP20. Another tumour suppressor that studies showed that UCHL1 expression stimulated has been linked to DUB activity is the von Hippel– oncogenesis and an invasive phenotype117–119, whereas Lindau (VHL) protein, which ubiquitylates hypoxia-­ UCHL1 depletion had antitumour effects and blocked inducible factor 1α (HIF1α) when cellular oxygen cell migration in a lung cancer cell line117. The precise levels are normal, leading to the degradation of HIF1α. mechanism by which UCHL1 contributes to tumor‑ USP20 (also known as VDU2), is reported to de­ubiqu­ igenesis remains unclear, although reports suggest itylate a number of proteins, including HIF1α. that it contributes to cell survival signalling, cell cycle

66 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

regulation, DNA repair and regulating pools of free offer promise for anticancer immunotherapies. In this ubiquitin in ways that affect protein degradation and regard, Mission Therapeutics is investigating USP7 as function115. UCHL1 inhibitors have been described, an immuno-oncology target and has developed USP7 the most potent being isatin O-acyl oximes (such as inhibitors (See Mission Therapeutics pipeline in Further LDN‑57444; TABLE 2) with some selectivity over UCHL3 information). (REF. 120). In addition, a series of pyridinones have been identified as moderate UCHL1 inhibitors121. Enzyme DUBs in neurodegenerative disease kinetic studies revealed that these compounds are Identification of ubiquitin in protein aggregates asso‑ uncompetitive inhibitors and are selective for UCHL1, ciated with neurodegenerative pathologies — such exhibiting no inhibition of other cysteine hydrolases as neurofibrillary tangles in Alzheimer disease, neu‑ tested. Using X-ray crystallography, a weak tripeptide ronal inclusion bodies (also known as Lewy bodies) fluoromethyl ketone inhibitor was subsequently shown in Parkinson disease or intranuclear inclusions in to bind within the UCHL1 , irreversibly mod‑ hereditary polyglutamine expansion disorders — has ifying the active-site cysteine122. Mission Therapeutics prompted much interest in understanding how ubiqui‑ has also developed several series of potent and selective tylation and deubiquitylation affect such aggregates134. UCHL1 inhibitors123,124. Although no UCHL1 inhibi‑ DUB function in the central nervous system has been tors have demonstrated antitumour activity in vivo, described in detail elsewhere135,136, therefore, below inducible depletion of UCHL1 has been shown to cause we focus on a selected number of DUBs connected to disease regression in an orthotopic multiple myeloma neurodegenerative disease. mouse model125. Mitochondrial quality control USP22. Another DUB associated with oncogenesis is Mitochondrial dysfunction and UPS impairment have USP22, the catalytic subunit of a deubiquitylase module been described as hallmarks of ageing137 and have been in the SAGA (Spt-Ada‑Gcn5‑acetyltransferase) complex. implicated in the aetiopathogenesis of many age-­related The best-characterized substrates for the SAGA com‑ diseases, particularly neurodegenerative disorders plex include several acetylation sites in histone H3 and such as Alzheimer disease and Parkinson disease. In a ubiquitylation site in histone H2B, post-­translational accord with this connection, ubiquitylation has close modification of which regulates gene expression29. links to mitochondrial function, with the UPS main‑ USP22 has strong links to oncogenesis29, having been taining mitochondrial homeostasis by regulating identified in microarray screens as part of an 11‑gene organelle dynamics, the mitochondrial proteome and ‘death-from-cancer’ signature for highly aggressive, mitophagy138. Conversely, mitochondrial dysfunction therapy-­resistant tumours. USP22 was later shown to act can impair cellular protein homeostasis by generating as an oncogene product, regulating cell cycle progres‑ oxidative damage. Notably, mutations in the ubiquitin sion, proliferation and apoptosis126. Increased expression E3 ligase Parkin are causally associated with certain cases of USP22 has been connected with poor prognosis in of familial Parkinson disease139. As Parkin ubiquitylates several cancers, including liver127, colorectal127 and breast mitochondrial components, thus promoting turnover cancers128 as well as oesophageal squamous cell carci‑ of mitochondria by lysosome-mediated mitophagy, defec‑ noma129 and oral squamous cell carcinoma130. If USP22 tive mitophagy and accumulation of defective mitochon‑ DUB activity can be linked to survival and progression dria that cause enhanced oxidative stress could be an of these cancers, then inhibitors may provide attractive underlying cause of Parkinson disease140,141. A corollary prospects for new therapies. of this is that Parkin activation — or inhibition of factors counteracting Parkin — could provide opportunities for Cancer immunotherapy disease alleviation. Given the role of ubiquitin modifications and DUBs in A screen for DUBs that oppose Parkin function iden‑ many inflammatory processes (see below), as well as the tified the mitochondrion-associated DUB USP30 as an renewed interest in targeting the immune system to fight antagonist of Parkin-mediated mitophagy142,143, with Regulatory T cells cancer, the antineoplastic potential of therapeutically USP30 depletion significantly decreasing mitochondrial (T cells). A subpopulation of reg inhibiting DUBs involved in the immune system is being numbers in cells, a phenotype that was rescued by wild- T cells that modulate the investigated. Among these is USP7, which positively reg‑ type but not catalytically inactive USP30. Furthermore, immune system, maintain tolerance to self-antigens and ulates the stability of FOXP3, a crucial transcription fac‑ USP30 depletion in vivo provided stress protection in prevent . tor controlling the differentiation of regulatory T cells (Treg Drosophila melanogaster models of Parkinson disease. cells)34. In a search for DUBs that contribute to GATA3 In line with such findings, USP30 depletion in human Mitochondria stabilization in FOXP3‑expressing cells, both USP7 and HeLa cells led to elongated and interconnected mito‑ Cellular organelles that 144 produce ATP through oxidative USP21 were shown to upregulate GATA3‑mediated chondria , suggesting a role for USP30 in regulating 131 phosphorylation. activity using a reporter assay . Furthermore, deple‑ mitochondrial fusion and fission. Current models invoke

tion of USP21 in Treg cells resulted in downregulation USP30 functioning under normal physio­logical condi‑ Mitophagy of FOXP3, compromised expression of Treg signature tions to prevent inappropriate mitophagy. However, in The process of eliminating genes and impaired their suppressive activity132. As T response to stresses such as membrane depolarization, damaged and aged reg mitochondria via sequestration cells restrict antitumour immune responses and promote Parkin is recruited to mitochondria to promote mito‑ 133 145 and hydrolytic degradation tumour survival , these results suggest that depletion phagy . Accordingly, under conditions of mitochon‑ by the lysosome. of FOXP3 in Treg cells by targeting USP7 and USP21 drial dysfunction — for example, caused by defects in

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 67 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

Parkin (or its positive regulator PINK1) — USP30 is overexpressed proteins, such as Tau, TAR DNA-binding thought to counteract clearance of damaged mitochon‑ protein 43 (TDP43) and ATXN3, whose accumulation dria, leading to a build‑up of metabolically and ener‑ is linked to neurodegenerative diseases46. Notably, IU1 getically deficient cells142. It is thus hypothesized that, only promoted degradation in Usp14+/+ murine embry‑ in the context of certain mitochondrial dysfunctions, onic fibroblasts46 but not in Usp14−/− cells, suggesting USP30 inhibition would have therapeutic benefits. that this compound functions specifically through So far, only one chemical inhibitor of USP30 has been USP14. Furthermore, IU1 reduced accumulation of described, 15‑oxospiramilactone (TABLE 2), which induced menadione-induced oxidized proteins and ameliorated mitochondrial elongation in Mfn1‑knockout mouse menadione- or hydrogen peroxide-induced cell death fibroblasts, with no effect on cell viability146. Mission in human HEK293 cells46. Proteostasis Therapeutics Therapeutics is exploring USP30 inhibition for the (in collaboration with Biogen) is developing USP14 treatment of Parkinson disease and other mitochondrial inhibitors for the clearance of aggregation-prone pro‑ disorders, and has published several patent applications teins, including α‑synuclein in Parkinson disease and describing USP30 inhibitors124,147. Tau in Alzheimer disease (see Proteostasis pipeline in Two other DUBs connected to mitophagy are USP8 Further information) and has published several patent and USP15. Notably, USP8 depletion was found to delay applications describing USP14 inhibitors153–155. Parkin translocation onto depolarized mitochondria, as Despite the growing interest in USP14 as a therapeu‑ well as mitochondrial clearance, and USP8 displayed tic target in cancer and neurodegeneration, the fact that an ability to remove K6‑linked ubiquitin chains from its loss causes severe morbidity and postnatal lethality Parkin in vitro148. In addition, USP8 has been shown to requires further investigation, especially in regard to remove K63‑linked ubiquitin chains from α-synuclein149, its role in neuromuscular junctions: the neuromuscu‑ a protein known to aggregate — often in a ubiquitylated lar phenotype of USP14‑deficient axJ mice is rescued form — in Lewy bodies associated with neurodegen‑ by neuronal-specific expression of USP14 (REF. 156). erative diseases such as Parkinson disease. Depletion Furthermore, the extent to which USP14 contributes to of USP8 in either human SH‑SY5Y cells or D. melano- the clearance of proteins involved in neurodegeneration gaster resulted in increased lysosomal degradation of in vivo remains controversial157. The development and α-synuclein149. Meanwhile, USP15 was identified as a use of USP14 inhibitors in disease-relevant models may Parkin-interacting protein that colocalizes with mito‑ shed further light on such issues and hopefully will define chondria150. In HeLa cells overexpressing Parkin, over‑ potential therapeutic windows for USP14 inhibition in expression of wild-type but not catalytically dead USP15 disease settings. strongly inhibited mitophagy143,150. Furthermore, deplet‑ ing endogenous USP15 enhanced mitophagy in HeLa USP16 cells, in a human dopaminergic neuronal cell line and Down syndrome is a congenital disorder driven by trip‑ in primary fibroblasts from human patients150. USP15 lication of human 21, on which the USP16 does not deubiquitylate Parkin under basal conditions gene resides. USP16 has been reported to regulate cell or when cells are treated with mitochondrial depolar‑ cycle progression and through deubiqu­ izing agents. USP15 depletion also does not seem to itylation of histone H2A158. Defects in haematopoietic affect Parkin translocation to mitochondria150, although stem cell self-renewal in a mouse model of Down syn‑ it can oppose Parkin-mediated mitochondrial ubiqui‑ drome were rescued by reducing USP16 expression to tylation. Finally, loss of USP15 in D. melanogaster was levels similar to those in control mice159. In addition, found to rescue both locomotor defects and accumu‑ USP16 overexpression in normal human fibroblasts lation of dysfunctional mitochondria in flight muscles and neural progenitors led to reduced cell expansion159, of parkin-knockout flies150. Collectively, these findings which was similar to the strong proliferation defects highlight the potential for USP8 and USP15 inhibitors in observed in fibroblasts from individuals with Down Parkinson disease and perhaps other diseases associated syndrome160. Thus, USP16 is a key regulator that con‑ with mitochondrial dysfunction. trols stem cell self-renewal and senescence in Down Further highlighting connections between Parkinson syndrome, suggesting that inhibitors of USP16 might disease and DUBs, ATXN3 has been shown to interact provide therapeutic benefits to such individuals. with Parkin in a manner that counteracts Parkin auto­ ubiquitylation151. In addition, USP7 was recently shown to DUBs in immunity and inflammation remove K63‑linked ubiquitin chains from α-­synuclein149, Pathogens are recognized by several families of pattern a protein that aggregates and accumulates in Lewy bodies, recognition receptors and activate various signal trans‑ which are hallmarks of Parkinson disease. duction cascades via the RIG-I-like receptors (RLRs), the nucleotide-binding oligomerization domain-like recep‑ 161 axJ USP14 tors (NLRs) and the toll-like receptors (TLRs) . These A hypomorphic allele of As described above, USP14 removes ubiquitin from cer‑ signalling events mediate induction of inflammation USP14. The ataxia mutation tain substrates targeted to the proteasome, thus rescu‑ that is important for recruiting immune cells to sites of J (ax ) is a recessive neurological ing such substrates from degradation and maintaining infection. Ubiquitylation is a critical post-­translational mutation that results in 54,152 (TABLE 2) 161 reduced growth, ataxia and a pool of free ubiquitin . IU1 , a reversible modification in this process . Non-degradative hindlimb muscle wasting small-molecule USP14 inhibitor, was shown to target the K63‑linked and M1‑linked ubiquitin chains mediate in mice. USP14 catalytic site and promote degradation of several the key upstream event of recruiting the TGFβ-activated

68 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

kinase 1 (TAK1) and the IκB kinase (IKK) complexes, human OTULIN has recently been shown to cause a respectively162. K63‑linked polyubiquitylation activates potentially fatal auto-inflammatory condition termed the TAK1 complex, which phosphorylates the sub­ OTULIN-related autoinflammatory syndrome183. unit IKKβ at key serine residues in the activation loop, Similar to ubiquitin, the UBL IFN-stimulated gene 15 resulting in IKK activation and transcriptional activation (ISG15) has a key role in cellular signalling in response of target genes that include mediators of immune and to pathogens. Conjugation of ISG15 to various cellular inflammatory responses as well as feedback inhibitors substrates is reversed by the IFN-inducible isopepti‑ of the NF‑κB pathway163. Negative regulators include dase USP18. USP18 is upregulated after viral infection, DUBs that cleave K63‑linked and linear ubiquitin type I and type III IFNs, lipopolysaccharide, TNFα or chains such as A20 (also known as TNFAIP3), CYLD genotoxic stress. In addition to its isopeptidase activ‑ and OTU domain-containing deubiquitinase with linear ity, USP18 negatively regulates type I and type III IFN linkage specificity (OTULIN; also known as FAM105B signalling by blocking the IFNAR2 subunit of the and Gumby)161,164,165. interferon receptor184. A20 is probably the best-characterized DUB linked to inflammation166. This DUB plays a key part in restrict‑ Inflammatory and autoimmune disorders ing TLR signalling and maintaining immune homeo­ Debilitating autoimmune diseases range from those with stasis through deubiquitylation of NF‑κB signalling genetic components — such as Crohn’s disease, diabetes factors such as NF-κB essential modulator (NEMO), mellitus type 1, Graves disease and rheumatoid arthritis185 receptor-interacting serine/threonine-protein kinase 1 — to sporadic conditions, including coeliac disease, (RIPK1) and tumour necrosis factor (TNF) receptor-­ inflammatory bowel disease, multiple sclerosis, psoriasis associated factor 6 (TRAF6)167. In addition, A20 can and systemic erythematosus. In addition, chronic bind to polyubiquitin chains through its zinc finger inflammatory diseases are characterized by a prolonged domain, allowing for interaction with ubiquitylated and persistent pro-inflammatory state, and include auto‑ NEMO. This ubiquitin-induced recruitment of A20 immune disease as well as metabolic syndrome, neuro‑ to NEMO is sufficient to block IKK phosphoryla‑ degenerative disease, chronic obstructive pulmonary tion by its upstream kinase TAK1, preventing NF‑κB disease and cardiovascular disease. activation168. Thus, A20 deficiency promotes local or Following pattern recognition receptor stimulation, systemic inflammation in vivo, underscoring why inac‑ dendritic cells secrete various cytokines that regulate tivating TNFAIP3 mutations have connections with both the differentiation of CD4+ T cells to different subsets 169 inflammatory and autoimmune syndromes . of T helper cells (TH cells), including inducible Treg cells,

CYLD is another DUB known to negatively regulate T follicular helper cells, and TH1, TH2, TH9 and TH17 186 ubiquitylation of RIG1 (one of the major RLRs) and cells . TH17 cells mediate pro-inflammatory functions RIG1‑mediated interferon (IFN)-regulated gene induc‑ through the secretion of pro-inflammatory cytokines, tion170,171. CYLD binds to RIG1 and thereby inhibits including interleukin‑17A (IL‑17A), IL‑17F, and IL‑22 (REF. 187) ubiquitylation and signalling functions of RIG1. CYLD . Moreover, TH17 cells have been implicated in also inhibits the ubiquitylation of TBK1 and IKKε, which the development of autoimmune diseases such as mul‑ contributes to the negative regulation of IFN responses171. tiple sclerosis, and systemic lupus Consistent with this, CYLD deficiency causes constitutive erythematosus188. activation of TBK1 and IKKε in dendritic cells. Despite USP4 has been shown to stabilize the nuclear recep‑ enhanced RIG1 signalling, CYLD-deficient cells and tor retinoid-related orphan receptor-γt (RORγt) in

mice are more susceptible to vesicular stomatitis virus TH17‑activated T cells and has been proposed as a pos‑ infection due to attenuated signalling and antiviral gene sible therapeutic target for rheumatoid arthritis189. One

expression induced by IFNβ, suggesting a positive role for report showed that USP4 is highly expressed in TH17 CYLD in the regulation of type I IFN receptor function161. cells, and its depletion resulted in decreased RORγt M1‑linked ubiquitin chains are generated by the as well as IL‑17A expression189. In addition, use of the linear ubiquitin chain assembly complex (LUBAC), reported USP4 inhibitor vialinin A (TABLE 2) also dimin‑ Cytokines which consists of haem-oxidized IRP2 ubiquitin ligase 1 ished RORγt and IL‑17A expression190. Furthermore, Members of a class of (HOIL1; also known as RBCK1), HOIL1‑interacting expression of USP4, IL‑17A and IL‑17F mRNA have been immunoregulatory proteins protein (HOIP; also known as RNF31) and SHANK- shown to be significantly elevated in CD4+ T cells from (interleukins or interferons) that are secreted by cells, associated RH domain-interacting protein (SHARPIN). patients with rheumatoid arthritis compared with healthy 189 especially cells of the LUBAC is recruited to many immune receptors and controls , providing further evidence for a role of USP4 immune system. ubiquitylates target proteins, including RIPK1, RIPK2, in rheumatoid arthritis. MYD88, interleukin‑1 receptor-associated kinases TRABID (also known as ZRANB1) is required for T helper cells 172,173 (T cells). Activated T cells (IRAKs) and NEMO . Genetic loss of LUBAC com‑ TLR-mediated expression of the inflammatory cytokines H 174 191 linked to defence against ponents leads to immunodeficiency and inflamma‑ IL‑12 and IL‑23 in dendritic cells . It has been proposed various types of pathogens, tory phenotypes in mice175–178, and mutations in LUBAC that TRABID deubiquitylates and stabilizes the histone which are commonly defined components also cause inflammatory conditions in Jumonji domain-containing protein 2D by their production of humans179,180. Hence, loss of M1‑linked ubiquitin chains (JMJD2D; also known as KDM4D), which regulates particular cytokines. The best-characterized effector imbalances immune signalling. OTULIN is the only histone modification at the Il12 and Il23 promoters 181,182 T cell subsets include TH1, TH2 DUB known to specifically cleave M1 linkages . to facilitate recruitment of the NF‑κB family member 191 and TH17 cells. Accordingly, a homozygous hypomorphic mutation in REL . Conditional deletion of Zranb1 in dendritic cells

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 69 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

impairs IL‑12 and IL‑23 production and the generation Viral infections

of TH1 and TH17 subsets of inflammatory T cells, ren‑ Ubiquitylation is important for modulation of pro‑ dering mice refractory to the induction of experimental tein–protein interactions, including the activation of autoimmune encephalomyelitis (EAE)191. innate immune signalling pathways, so perhaps it is not

Another DUB associated with the activity of TH17 surprising that various viruses have genes for DUBs, cells is USP18. Although this DUB has been extensively which are used as a strategy to inhibit ubiquitin- and studied in the context of viral infection, USP18 regu‑ ISG15‑dependent antiviral pathways197. Severe acute lates the TAK1–TAB interaction, which is required for respiratory syndrome coronavirus (SARS-CoV) and 192 TH17 cell differentiation and the autoimmune response . Middle East respiratory syndrome coronavirus (MERS- Consistent with this, USP18‑deficient mice were resistant CoV) are two of the six known human coronaviruses. to EAE192. Both are highly pathogenic, with the potential for T cell receptor signalling has been shown to be facili‑ human‑to‑human transmission, and contain papain-like tated by the DUB cellular zinc finger anti-NF-κB protein cysteine proteases termed SARS-CoV PLpro and MERS- (CEZANNE; also known as OTUD7B), which binds to CoV PLpro, respectively. In addition to processing viral and deubiquitylates ζ-chain-associated protein (ZAP70), polyprotein, these proteases remove ubiquitin and ISG15 thus preventing the interaction of ZAP70 with negative from host cell factors, resulting in antagonism of the host regulatory phosphatases193. ZAP70 is a cyto­plasmic antiviral immune response198. Hence, both SARS-CoV protein tyrosine kinase that plays a crucial part in T cell PLpro and MERS-CoV PLpro have been proposed as signalling and is recruited to phosphorylated sites on important antiviral targets. The X‑ray structures of both the T cell receptor, where it is subsequently phospho‑ proteases have shown structural similarity to the USP rylated by the SRC kinase LCK. Phosphorylation of family of DUBs199–201. ZAP70 is required for full activation and downstream OTU domain-containing proteases from diverse phosphorylation of adaptor proteins, which facilitate RNA viruses, including the nairoviruses Crimean- T cell signalling194. In addition, CEZANNE-deficient Congo haemorrhagic fever virus and Dugbe virus, the mice exhibited attenuated T cell responses to bacterial papain-­like protease (PLP2) domain of the arterivirus infection and were refractory to EAE193. While young equine arteritis virus, and the protease (PRO) domain of Otud7b‑knockout mice had similar naive and memory-­ the tymovirus turnip yellow mosaic virus can hydrolyse like T cells compared with wild-type mice, older mice ubiquitin­ and ISG15 from cellular target proteins197,202. deficient for CEZANNE had reduced IFNγ-producing Many positive-strand RNA viruses, including arterivi‑ 193 TH1 cell subsets . ruses and tymoviruses, encode polyproteins that are post-­

Similar to TH17 cells, TH1 cells have the capacity translationally cleaved by internal protease domains. In to cause inflammation and autoimmune disease. The accord with this, both arterivirus PLP2- and tymovirus

development, differentiation and function of TH1 cells PRO-domain-containing proteases are crucially required is driven by the T‑box transcriptional factor T‑bet (also for viral replication due to their primary role in polyprotein­ 197 known as TBX21), which promotes TH1-mediated maturation . Thus, viral OTU domain-containing immune response primarily through promoting expres‑ proteases may represent promising therapeutic targets. sion of the cytokine IFNγ195. The DUB USP10 has been shown to deubiquitylate and stabilize T‑bet, resulting Bacterial infections in enhanced secretion of IFNγ196. In addition, USP10 Bacteria use a repertoire of effector proteins that target the mRNA expression was found to be elevated in periph‑ eukaryotic ubiquitin system to promote bacterial patho‑ eral blood mononuclear cells from patients with asthma genicity. Protease activities from human bacterial patho‑ compared with healthy donors196. gens, including Salmonella enterica serovar Typhimurium Although it is currently unclear why so many DUBs (SseL), Escherichia coli (ElaD), Shigella flexneri (ShiCE), are involved in the regulation of immune responses, it Chlamydia trachomatis (ChlaDUB1), Rickettsia bellii is possible that different DUBs function in distinct cell (RickCE) and Legionella pneumophila subsp. pneu- types. Many published studies are based on cell lines mophila (LegCE), have recently been characterized203. and overexpression systems, and the expression of LegCE showed no proteolytic activity; SseL, ElaD, and endo­genous DUBs in various immune cells will be an ShiCE demonstrated ubiquitin-­specific protease activity; important area for future investigation. Similarly, the whereas ChlaDUB1 and RickCE cleaved both ubiqu­ generation of genetic models and the development of itin and, to a lesser extent, NEDD8‑modified peptides. inhibitors for CEZANNE, TRABID, USP4, USP10 and Interestingly, these DUBs encoded by human pathogens USP18 will help determine their therapeutic potential. showed strong preference for K63‑linked ubiquitin chains, only targeting K48‑linked and K11‑linked chains at later DUBs in infectious diseases time points or higher enzyme concentrations. Therefore, As described below, there is growing interest in DUBs as bacterial DUBs are potential therapeutic targets. potential therapeutic targets for various infectious dis‑ eases of man and other animals. Such potential is being Parasitic infections explored both by developing compounds that inhibit In addition to expressing DUBs that target host func‑ the activity of pathogen-encoded DUB-like proteins, or tions, similar to viruses and bacteria, eukaryotic para‑ target host cell DUBs that control the pathogen life cycle sites also possess UBL pathways of their own. The use or infectivity. of ubiquitin-based activity probes to identify DUBs

70 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

in Plasmodium falciparum led to the identification proteolytic reaction between a lysine ε-side chain and a of PfUCH54, which was shown to have deubiquityl‑ carboxyl group corresponding to the ubiquitin C termi‑ ating activity and also an ability to remove adducts nus209. The last two C‑terminal amino acid residues are of the UBL NEDD8 (REF. 204). Further investigation of glycines (Gly75–Gly76) that lack side chains, resulting the parasite Toxoplasma gondii using a similar strategy in a narrow linker on either side of the isopeptide bond, identified four DUBs, one of which was orthologous to which is mirrored in a long and narrow DUB catalytic mammalian UCHL3 (REF. 205). Structural studies on cleft209. Moreover, cysteinyl protease DUB catalytic activ‑ PfUCHL3 explained the dual specificity of the enzyme, ity tends to rely on two or three crucial residues compris‑ and PfUCHL3 was found to be required for parasite sur‑ ing a catalytic diad or triad, generally containing a His 206 vival . Distinct differences in the ubiquitin- side chain that — by lowering the pKa of the catalytic between PfUCHL3 and its human counterpart suggest Cys — leads to a nucleophilic attack on the ubiquitin– that this parasitic DUB can be selectively targeted by substrate isopeptide linkage12. Collectively, these prop‑ inhibitors. Based on the above findings, it will be of great erties bring complexity to the identification of selective interest to further explore anti-infective opportunities small-­molecule inhibitors that target DUB catalytic sites for DUB inhibitors. and are likely to restrict the breadth of series that are suit‑ able for developing potent and selective DUB inhibitors. Challenges and emerging technologies The Proteostasis thiophene pyrimidine core-derived Despite the significant and growing attractiveness of USP14 inhibitors are known to bind in the ubiquitin DUBs as drug targets, DUB-focused drug discovery pocket and prevent the ubiquitylated substrate from has been challenging, with researchers in this arena binding210. However, the majority of historical and current facing various obstacles. First, although DUBs have DUB drug discovery programmes have focused on chem‑ clear catalytic pockets that a priori seem suitable for ical series that include the provision of an active ‘warhead’ drug development, a key challenge has been to identify that forms a reversible or irreversible covalent adduct with potent compounds that show selectivity among related the DUB catalytic cysteine. The high reactivity of some of DUBs and that have properties commensurate with their these warheads, which include oxidative, alkylating and development for clinical use. Second, ubiquitylation and arylating moieties210, is likely to limit drug selectivity, may deubiquitylation are intracellular processes that, at least hamper the development of acceptable pharmacokinetic at present, are only amenable to classical small-­molecule and pharmacodynamic parameters, and may also pose chemical approaches. Third, because most DUBs exe‑ risks of idiosyncratic toxicities in patients. cute the transfer of ubiquitin molecules via a reactive For this reason, less-reactive warheads are being thiol group, most standard assays used to identify explored that are closely related to warheads utilized by inhibitors are prone to non-selective redox or alkylating non-DUB cysteine protease inhibitors in the clinic. For false positives207. Fourth, the mechanisms of action of example, the USP8 inhibitor identified from a library of DUB enzymes are often complex, involving regulation amidomethyl methyl acrylates (Compound 6)211 contains of enzymatic activity through allosteric effects and/or a Michael acceptor group also found in rupintrivir, an substrate-mediated catalysis, and many DUBs alter‑ inhibitor of rhinovirus 3C protease and a GSK cathepsin nate between active and non-active conformations (see C inhibitor210. In addition, USP9X inhibitors WP1130 below)208,209. This makes it challenging both to design and EOA1342143 (REF. 212) contain a Michael acceptor predictive biochemical assays and develop drug-like group similar to that found in a Principia Biopharma’s compounds. Finally, DUBs often display specificity for Bruton tyrosine kinase (BTK) inhibitor210. However, ubiquitin chains as well as the target proteins. Hence, to these examples are few in number, and the compounds optimize the likelihood of identifying genuine inhibitors, are weak DUB inhibitors. Mission Therapeutics has it is prudent to develop bespoke primary screening and discovered covalent active-site series that are ‘drug-like’, secondary assays that recapitulate the most physiological unrelated to any previously described DUB inhibitor, and substrate and ubiquitin-linkage setting for each DUB. that achieve submicromolar cell-based potencies Despite the above issues, DUBs are fundamentally and exhibit good oral bioavailability123,124,147. catalytically driven proteins with known enzymatic functions and, as such, present researchers with the : implications opportunity to identify small-molecule inhibitors either Most peptidases, including many cysteine proteases, within the active site or at adjacent allosteric pockets. recognize a small linear-polypeptide motif and cleave Indeed, over the past few years there has been an increas‑ either before or after the peptide bond213. DUBs, how‑ ing rate of progress in successfully screening for and ever, are more complex. Most DUBs cleave an isopeptide evolving small-molecule DUB inhibitors, with the most linkage between the side chain of a lysine residue and developed ones now moving towards or into clinical the C-terminal glycine of ubiquitin, with the isopep‑ evaluation (for examples, see TABLE 2). tide linkage providing specificity and flexibility to the mechanism of proteolysis214. Also, DUBs need to accom‑ Understanding DUB–substrate interactions modate substantial globular post-translational modifi‑ Allosteric Understanding the mechanism of action of individ‑ cations (ubiquitin, UBL, or ubiquitin–UBL chains) into Binding to a site other than 215 the active site of the enzyme, ual DUBs is important when initiating any screening their catalytic site . Furthermore, unlike most other resulting in modulation of and subsequent drug discovery campaign. DUBs are cysteine peptidases, the of cysteinyl pepti‑ enzymatic activity. generally isopeptidases that, in most cases, catalyse a dase DUBs is not usually in a ‘functional’ configuration,

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 71 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

with allosteric regulation being required to render DUBs high false-positive rates, an issue that has likely been the fully functional and processive. Such allosteric regulation most significant challenge in identifying genuine and can be substrate-mediated (for example, OTULIN)181, selective DUB inhibitors. triggered by intramolecular reorganization (for example, Indeed, the non-selective nature of some DUB inhib‑ USP7)216 or induced by key cofactors (as for USP1)74. In itors is highlighted in biochemical selectivity-profiling addition, several DUBs are associated with multi-protein assays, with relatively few DUB inhibitors reported in complexes such as the proteasome217, valosin-containing the literature showing promise in such studies76. Matrix- protein (VCP; also known as p97)218, or the COP9 sig‑ assisted laser desorption/ionization time-of-flight nalosome219. These associations can allosterically regu‑ (MALDI-TOF) mass spectrometry was used to screen for late the affinity of DUBs for their substrates208,220, and in DUB activity and specificity by systematically assessing some instances, DUBs coexist in the same complex as the the specificities of 42 recombinant human DUBs against ubiqu­itylation machinery221. The above issues must there‑ di‑ubiquitin isomers with all possible chain linkages fore be carefully considered when establishing screening (M1- (linear), K6-, K11-, K27-, K29-, K33-, K48- and and compound evaluation assays for a DUB. Some DUB K63‑linked)76. Subsequently, a panel of 32 DUBs was inhibitors have been suggested to target allosteric sites, screened against 9 reported DUB inhibitors. Their find‑ such as the USP1 inhibitor ML323 (REF. 78). ings demonstrated that none of the compounds displayed strong selectivity towards a single DUB and that many Screening technologies inhibited most DUBs on the panel. Approximately twenty years ago, a general assay was Novel technologies based on chemically synthesized established for measuring DUB enzymatic activity DUB substrates containing isopeptide linkages, ubiquitin based on the substrate ubiquitin C‑terminal 7‑amido‑4‑ chains and/or assay technologies that are less prone to false methylcoumarin (Ub‑AMC). This substrate is efficiently positives — such as luminescence, time-resolved fluores‑ cleaved or hydrolysed by various DUBs, releasing a highly cence or mass spectrometry — are advancing screening fluorescent AMC moiety. While this assay has been used campaigns and are therefore now being exploited76,226–228. in various DUB inhibitor screens, for example to identify For example, a ubiquitin–aminoluciferin substrate was USP1 (REF. 222) and USP7 (REFS 207,223,224) inhibitors, used with a variety of DUBs to demonstrate a suitable assay one significant drawback is that it is prone to fluores‑ window for high-throughput screening207,229. Subsequently, cence interference exhibited by many small molecules225. USP2 was used as a representative DUB to demonstrate Moreover, AMC and alternative tags such as rhodamine statistical robustness of this reagent in a screening cam‑ and tetramethylrhodamine (TAMRA), which have been paign for inhibitors. We believe that such developments used because they are less prone to fluorescence artefacts, are crucial to optimize the prospects for identifying and contain a peptide linkage and thus differ quite substan‑ developing DUB inhibitors for ultimate clinical use. tially from most natural DUB substrates. Processing of such substrates thus requires the DUB to function in a Monitoring DUB activity or inhibition non-physiological manner, thereby potentially diminish‑ A key issue when studying DUBs and their modula‑ ing prospects for identifying compounds that will operate tion in cells is understanding substrate specificity. Some in cellular or therapeutic settings. DUBs have preferences for monoubiquitylated substrates, A further challenge for development of DUB inhibitor whereas others favour specific ubiquitin chain types, screening assays is oxidative hydrolysis of the active-site chains bearing mixed linkages, or mixed chains contain‑ cysteinyl residue of purified DUBs in biochemical buff‑ ing ubiquitin and UBLs230,231. Furthermore, many DUBs ers. This sensitivity requires use of protective reducing have some specificity for the substrate protein itself, with agents such as dithiothreitol (DTT), usually in milli­ this being mediated through mechanisms often involving molar concentrations, to maintain DUB enzymatic activ‑ regions of the DUB distinct from its catalytic site. DUB ity. Altering the concentration or type of reducing agent substrates can be determined by biochemistry, yeast (for example, 2‑mercaptoethanol, cysteine, glutathione or two‑hybrid interactions, proteomic profiling and genet‑ tris(2‑carboxyethyl)phosphine­ (TCEP)) can considerably ics232, but this is often challenging and time-consuming. affect inhibition obtained for hit compounds207. Following Clearly, the ability to directly monitor DUB activity within a high-throughput screen to identify USP7 inhibitors, the a native biological system is essential to understanding the ability of compounds to inhibit USP7 in the presence of physiological and pathological role of individual DUBs as different reductants was evaluated207. Many compounds well as the effects of DUB inhibition233. showed the greatest inhibition in the absence of any DUB activity in cells can be monitored by chemical reductant, being less potent in the presence of cysteine or probes that generate readily detectable covalent complexes glutathione, and least potent in the presence of DTT with the DUB catalytic site (recently reviewed in REF. 234). or TCEP. A further subset of molecules showed an alterna‑ Activity probes label DUBs based on their catalytic-site tive profile, only demonstrating inhibition in the presence thiol group235, with DUB reactivity towards such probes Activity-based probes of DTT or TCEP. A final set of molecules only inhibited depending on the type of electrophilic warhead fused to (ABPs). Tools that incorporate USP7 when no additional reductant was added. Together, ubiquitin. In addition to profiling DUB levels, activity elements for targeting, these data demonstrate the critical nature of the reducing and catalytic inhibition, activity probes have also been modification and/or detection of labelled proteins, enabling environment on DUB activity and inhibition. Thus, most used to identify DUBs by affinity purification combined 236 assessment of enzymatic screens based on high concentrations of reducing agents with mass spectrometry . More recently, activity-based activity and inhibition. and using first-generation fluorescent substrates generate probes (ABPs) bearing a fluorescent reporter tag have

72 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

been generated to replace the initial tags (for example, The limitation to this approach, however, is the number the haemagglutinin (HA) epitope) to allow fluorescent of enzymes successfully labelled by the ABP and the rep‑ imaging instead of detection by immuno­blotting226,227. resentation of active enzymes in the cellular proteome. Although production of ubiquitin ABPs was historically ABPs were used to characterize the DUB inhibitors based on a trypsin-catalysed transpeptidation to modify PR‑619 and P22077 by immuno­precipitation, combined ubiquitin at its C terminus with a vinyl sulfone group, with identification and label-free quantification by mass recent approaches involve the full chemical synthesis of spectrometry-based proteomics239. Using this approach, ubiquitin ABPs226,237. This advance allows incorporation quantitative data for 25 cellular DUBs was obtained. of modified amino acid residues at any position in the PR‑619 was confirmed as a broad DUB inhibitor, whereas ABPs, whether natural or not. Mass spectrometry has P22077 was found to be a selective inhibitor of USP7 and become an important tool to monitor ubi­quitin adducts USP47 that may therefore provide the basis for exploring as well as changes in ubiquitin levels232,238. Indeed, com‑ therapeutic opportunities in oncology (see preceding sec‑ bining ABPs with immunoblotting or mass spectrometry tions and TABLE 2). can generate powerful tools for monitoring DUB activity and inhibition by small molecules98,239 as well as assessing Concluding remarks drug–enzyme target engagement in cells or tissues. For During the past decade, we have witnessed dramatic example, one study used ABPs to demonstrate the selec‑ advances in our understanding of DUB functions, mecha‑ tivity of P22077 for USP7 in cells, in contrast to PR‑619, nisms of action, regulation and disease linkages. In parallel, which inhibited a broad range of DUBs239. In addition, there have been major improvements in DUB biochemical another study demonstrated the cellular selectivity of assays and screening technologies, leading to the devel‑ HBX19818 for USP7 against a panel of DUBs using ABPs opment of increasing numbers of small-­molecule DUB and immunoblotting98. inhibitors whose selectivity is now being explored and, Activity-based proteomic probes have facilitated the where possible, refined. Such inhibitors are providing the development of pharmacologically active enzyme inhibi‑ basis for drug-like molecules suitable for clinical evaluation tors. This approach represents a cell-based assay in which and are also providing versatile tools to further investigate treatment with the inhibitor is carried out in intact cells, DUB cell biology, regulation and biochemical mecha‑ allowing for a range of cellular enzymes to be assessed nisms, as well as to test therapeutic hypotheses in disease simultaneously239. Competition assays between an inhibi‑ models. Although it is still too early to predict the extent of tor and the ABP lead to a reduced labelling profile for the the broad therapeutic potential of DUBs, the next few years ABP, with loss of signal for ABP-labelled target enzymes certainly seem set to produce further exciting developments allowing for assessment of the specificity of inhibition. in the arenas of DUB biology and drug discovery.

1. Ciechanover, A. The ubiquitin proteolytic system and 13. van der Veen, A. G. & Ploegh, H. L. Ubiquitin-like 24. Kawaguchi, Y. et al. CAG expansions in a novel gene pathogenesis of human diseases: a novel platform for proteins. Annu. Rev. Biochem. 81, 323–357 (2012). for Machado-Joseph disease at chromosome mechanism-based drug targeting. Biochem. Soc. 14. Huang, C. J., Wu, D., Khan, A. F.. & Huo, L. J. 14q32.1. Nat. Genet. 8, 221–228 (1994). Trans. 31, 474–481 (2003). DeSUMOylation: an important therapeutic target and 25. McDonell, L. M. et al. Mutations in STAMBP, encoding 2. Ciechanover, A. : from the lysosome to protein regulatory event. DNA Cell Biol. 34, 652–660 a , cause microcephaly- ubiquitin and the proteasome. Nat. Rev. Mol. Cell Biol. (2015). capillary malformation syndrome. Nat. Genet. 45, 6, 79–87 (2005). 15. Murali, R., Wiesner, T. & Scolyer, R. A. Tumours 556–562 (2013). 3. Gallastegui, N. & Groll, M. The 26S proteasome: associated with BAP1 mutations. Pathology 45, 26. Cohen, P. & Tcherpakov, M. Will the ubiquitin system assembly and function of a destructive machine. 116–126 (2013). furnish as many drug targets as protein kinases? Cell Trends Biochem. Sci. 35, 634–642 (2010). 16. Oliveira, A. M. & Chou, M. M. USP6‑induced 143, 686–693 (2010). 4. Finley, D., Chen, X. & Walters, K. J. Gates, channels, neoplasms: the biologic spectrum of aneurysmal bone This is a key perspective highlighting DUBs as and switches: elements of the proteasome machine. cyst and nodular fasciitis. Hum. Pathol. 45, 1–11 attractive drug targets. Trends Biochem. Sci. 41, 77–93 (2016). (2014). 27. Huang, X. & Dixit, V. M. Drugging the undruggables: 5. Peth, A., Besche, H. C. & Goldberg, A. L. 17. Hao, Y. H. et al. USP7 acts as a molecular rheostat to exploring the ubiquitin system for drug development. Ubiquitinated proteins activate the proteasome by promote WASH-dependent endosomal protein recycling Cell Res. 26, 484–498 (2016). binding to Usp14/Ubp6, which causes 20S gate and is mutated in a human neurodevelopmental This recent review highlights drug discovery efforts opening. Mol. Cell 36, 794–804 (2009). disorder. Mol. Cell 59, 956–969 (2015). targeting the ubiquitin system, including DUBs. 6. Muratani, M. & Tansey, W. P. How the ubiquitin- 18. Ma, Z. Y. et al. Recurrent gain-of-function USP8 28. Sacco, J. J., Coulson, J. M., Clague, M. J. & Urbe, S. proteasome system controls transcription. Nat. Rev. mutations in Cushing’s disease. Cell Res. 25, Emerging roles of deubiquitinases in cancer-associated Mol. Cell Biol. 4, 192–201 (2003). 306–317 (2015). pathways. IUBMB Life 62, 140–157 (2010). 7. Jesenberger, V. & Jentsch, S. Deadly encounter: 19. Reincke, M. et al. Mutations in the deubiquitinase 29. Wang, L. & Dent, S. Y. Functions of SAGA in develo- ubiquitin meets apoptosis. Nat. Rev. Mol. Cell Biol. 3, gene USP8 cause Cushing’s disease. Nat. Genet. 47, pment and disease. Epigenomics 6, 329–339 (2014). 112–121 (2002). 31–38 (2015). 30. Nicholson, B. & Suresh Kumar, K. G. The multifaceted 8. Hicke, L. Protein regulation by monoubiquitin. Nat. References 18 and 19 are key publications roles of USP7: new therapeutic opportunities. Reviews Molecular Cell Biology 2, 195–201 (2001). identifying mutations in USP8 that enhance EGFR Cell Biochem. Biophys. 60, 61–68 (2011). 9. Jackson, S. P. & Durocher, D. Regulation of DNA signalling and cause Cushing disease. 31. Cremona, C. A., Sancho, R., Diefenbacher, M. E. & damage responses by ubiquitin and SUMO. Mol. Cell 20. Homan, C. C. et al. Mutations in USP9X are Behrens, A. Fbw7 and its counteracting forces in stem 49, 795–807 (2013). associated with X-linked intellectual disability and cells and cancer: Oncoproteins in the balance. This review highlights how ubiquitylation and disrupt neuronal cell migration and growth. Semin. Cancer Biol. 36, 52–61 (2016). related processes control many cellular responses Am. J. Hum. Genet. 94, 470–478 (2014). 32. D’Arcy, P., Wang, X. & Linder, S. Deubiquitinase to DNA damage. 21. Murtaza, M., Jolly, L. A., Gecz, J. & Wood, S. A. inhibition as a cancer therapeutic strategy. Pharmacol. 10. Husnjak, K. & Dikic, I. Ubiquitin-binding proteins: La FAM fatale: USP9X in development and disease. Ther. 147, 32–54 (2015). decoders of ubiquitin-mediated cellular functions. Cell. Mol. Life Sci. 72, 2075–2089 (2015). 33. Souroullas, G. P. & Sharpless, N. E. Stem cells: Down’s Annu. Rev. Biochem. 81, 291–322 (2012). 22. Eichhorn, P. J. et al. USP15 stabilizes TGF-beta syndrome link to ageing. Nature 501, 325–326 (2013). 11. Herhaus, L. & Dikic, I. Expanding the ubiquitin code receptor I and promotes oncogenesis through the 34. van Loosdregt, J. et al. Stabilization of the through post-translational modification. EMBO Rep. activation of TGF-beta signaling in glioblastoma. transcription factor Foxp3 by the deubiquitinase USP7 16, 1071–1083 (2015). Nat. Med. 18, 429–435 (2012). increases Treg-cell-suppressive capacity. Immunity 39, 12. Komander, D., Clague, M. J. & Urbe, S. Breaking the 23. Bignell, G. R. et al. Identification of the familial 259–271 (2013). chains: structure and function of the deubiquitinases. cylindromatosis tumour-suppressor gene. Nat. Genet. First report demonstrating that USP7

Nat. Rev. Mol. Cell Biol. 10, 550–563 (2009). 25, 160–165 (2000). deubiquitylates and stabilizes FOXP3 in Treg cells.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 73 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

35. Sun, H. et al. Bcr-Abl ubiquitination and Usp9x 62. Tian, Z. et al. A novel small molecule inhibitor of 86. Kapuria, V. et al. Deubiquitinase inhibition by small- inhibition block kinase signaling and promote CML deubiquitylating enzyme USP14 and UCHL5 induces molecule WP1130 triggers formation and cell apoptosis. Blood 117, 3151–3162 (2011). apoptosis in multiple myeloma and overcomes tumor cell apoptosis. Cancer Res. 70, 9265–9276 36. Savio, M. G. et al. USP9X controls EGFR fate by bortezomib resistance. Blood 123, 706–716 (2014). (2010). deubiquitinating the endocytic adaptor Eps15. Curr. 63. Zhou, H.‑J. et al. Compositions and methods for JAMM 87. Kapuria, V. et al. A novel small molecule deubiquitinase Biol. 26, 173–183 (2016). protein inhibition. WO2012158435 (A1) (2012). inhibitor blocks Jak2 signaling through Jak2 37. Richardson, P. G., Hideshima, T. & Anderson, K. C. 64. Zhou, H.‑J., Parlati, F. & Wustrow, D. Methods and ubiquitination. Cell Signal 23, 2076–2085 (2011). Bortezomib (PS‑341): a novel, first-in-class proteasome compositions for JAMM protease inhibition. 88. Clague, M. J. et al. Deubiquitylases from genes to inhibitor for the treatment of multiple myeloma and WO2013123071 (A1) (2013). organism. Physiol. Rev. 93, 1289–1315 (2013). other cancers. Cancer Control 10, 361–369 (2003). 65. Zhou, H.‑J. & Wustrow, D. Compositions and methods 89. Prives, C. Signaling to p53: breaking the MDM2 p53 38. Chen, D., Frezza, M., Schmitt, S., Kanwar, J. & for JAMM protein inhibition. WO2014066506 (A2) circuit. Cell 95, 5–8 (1998). Dou, Q. P. Bortezomib as the first proteasome inhibitor (2014). 90. Harris, S. L. & Levine, A. J. The p53 pathway: positive anticancer drug: current status and future perspectives. 66. O’Connor, M. J. Targeting the DNA Damage response and negative feedback loops. Oncogene 24, Curr. Cancer Drug Targets 11, 239–253 (2011). in cancer. Mol. Cell 60, 547–560 (2015). 2899–2908 (2005). 39. Yao, T. & Cohen, R. E. A cryptic protease couples 67. Jackson, S. P. & Bartek, J. The DNA-damage response 91. Li, M. et al. Deubiquitination of p53 by HAUSP is an deubiquitination and degradation by the proteasome. in human biology and disease. Nature 461, important pathway for p53 stabilization. Nature 416, Nature 419, 403–407 (2002). 1071–1078 (2009). 648–653 (2002). 40. Song, Y. et al. Deubiquitylating enzyme Rpn11/POH1/ 68. Jacq, X., Kemp, M., Martin, N. M. & Jackson, S. P. 92. Li, M., Brooks, C. L., Kon, N. & Gu, W. A dynamic role PSMD14 As therapeutic target in multiple myeloma. Deubiquitylating enzymes and DNA damage response of HAUSP in the p53‑Mdm2 pathway. Mol. Cell 13, Blood 128, 4469–4469 (2016). pathways. Cell Biochem. Biophys. 67, 25–43 (2013). 879–886 (2004). 41. Wang, B. et al. POH1 deubiquitylates and stabilizes 69. Castella, M. et al. FANCI Regulates recruitment of the 93. Cummins, J. M. & Vogelstein, B. HAUSP is required for E2F1 to promote tumour formation. Nat. Commun. 6, FA core complex at sites of DNA damage p53 destabilization. Cell Cycle 3, 689–692 (2004). 8704 (2015). independently of FANCD2. PLoS Genet. 11, 94. van der Horst, A. et al. FOXO4 transcriptional activity 42. Liu, H., Buus, R., Clague, M. J. & Urbe, S. Regulation e1005563 (2015). is regulated by monoubiquitination and USP7/HAUSP. of ErbB2 receptor status by the proteasomal DUB 70. Nijman, S. M. et al. The Deubiquitinating enzyme Nat. Cell Biol. 8, 1064–1073 (2006). POH1. PLoS ONE 4, e5544 (2009). USP1 regulates the Fanconi anemia pathway. 95. Song, M. S. et al. The deubiquitinylation and 43. Kakarougkas, A. et al. Co-operation of BRCA1 and Mol. Cell 17, 331–339 (2005). localization of PTEN are regulated by a HAUSP-PML POH1 relieves the barriers posed by 53BP1 and 71. Huang, T. T. et al. Regulation of monoubiquitinated network. Nature 455, 813–817 (2008). RAP80 to resection. Nucleic Acids Res. 41, PCNA by DUB autocleavage. Nat. Cell Biol. 8, 96. Lecona, E. et al. USP7 is a SUMO deubiquitinase 10298–10311 (2013). 339–347 (2006). essential for DNA replication. Nat. Struct. Mol. Biol. 44. Hu, M. et al. Structure and mechanisms of the 72. Guervilly, J. H., Renaud, E., Takata, M. & Rosselli, F. 23, 270–277 (2016). proteasome-associated deubiquitinating enzyme USP1 deubiquitinase maintains phosphorylated CHK1 97. Colland, F. et al. Small-molecule inhibitor of USP7/ USP14. EMBO J 24, 3747–3756 (2005). by limiting its DDB1‑dependent degradation. HAUSP ubiquitin protease stabilizes and activates p53 45. Eletr, Z. M. & Wilkinson, K. D. Regulation of Hum. Mol. Genet. 20, 2171–2181 (2011). in cells. Mol. Cancer Ther. 8, 2286–2295 (2009). proteolysis by human deubiquitinating enzymes. 73. Williams, S. A. et al. USP1 Deubiquitinates ID proteins 98. Reverdy, C. et al. Discovery of specific inhibitors of Biochim. Biophys. Acta 1843, 114–128 (2014). to preserve a mesenchymal stem cell program in human USP7/HAUSP deubiquitinating enzyme. 46. Lee, B. H. et al. Enhancement of proteasome activity osteosarcoma. Cell 146, 918–930 (2011). Chem. Biol. 19, 467–477 (2012). by a small-molecule inhibitor of USP14. Nature 467, 74. Cohn, M. A. et al. A UAF1‑containing multisubunit 99. Chauhan, D. et al. A small molecule inhibitor of 179–184 (2010). protein complex regulates the Fanconi anemia ubiquitin-specific protease‑7 induces apoptosis in 47. Wu, N. et al. Over-expression of deubiquitinating pathway. Mol. Cell 28, 786–797 (2007). multiple myeloma cells and overcomes bortezomib enzyme USP14 in lung adenocarcinoma promotes 75. Chen, J. et al. Selective and cell-active inhibitors of the resistance. Cancer Cell 22, 345–358 (2012). proliferation through the accumulation of beta- USP1/ UAF1 deubiquitinase complex reverse cisplatin 100. Fan, Y. H. et al. USP7 inhibitor P22077 inhibits catenin. Int. J. Mol. Sci. 14, 10749–10760 (2013). resistance in non-small cell lung cancer cells. neuroblastoma growth via inducing p53‑mediated 48. Wang, Y. et al. Ubiquitin-specific protease 14 (USP14) Chem. Biol. 18, 1390–1400 (2011). apoptosis. Cell Death Dis. 4, e867 (2013). regulates cellular proliferation and apoptosis in 76. Ritorto, M. S. et al. Screening of DUB activity and 101. Gavory, G. et al. Discovery and characterization of epithelial ovarian cancer. Med. Oncol. 32, 379 (2015). specificity by MALDI-TOF mass spectrometry. novel, highly potent and selective USP7 inhibitors. 49. Xu, D. et al. Phosphorylation and activation of Nat. Commun. 5, 4763 (2014). Cancer Res. 75 (Suppl. 15), abstr. LB‑257 (2015). ubiquitin-specific protease‑14 by Akt regulates the Key publication assessing substrate specificity of 102. Liu, H. et al. The Machado–Joseph disease ubiquitin-proteasome system. eLife 4, e10510 (2015). 42 DUBs and selectivity profiling of 9 reported deubiquitinase ataxin‑3 regulates the stability and 50. Jung, H. et al. Deubiquitination of Dishevelled by DUB inhibitors against a panel of 32 DUBs. apoptotic function of p53. PLoS Biol. 14, e2000733 Usp14 is required for Wnt signaling. Oncogenesis 2, 77. Dexheimer, T. S. et al. Synthesis and structure-activity (2016). e64 (2013). relationship studies of 103. Cuella-Martin, R. et al. 53BP1 integrates DNA repair 51. Qiu, X. B. et al. hRpn13/ADRM1/GP110 is a novel N-benzyl‑2‑phenylpyrimidin‑4‑amine derivatives as and p53‑dependent cell fate decisions via distinct proteasome subunit that binds the deubiquitinating potent USP1/UAF1 deubiquitinase inhibitors with mechanisms. Mol. Cell 64, 51–64 (2016). enzyme, UCH37. EMBO J. 25, 5742–5753 (2006). anticancer activity against nonsmall cell lung cancer. 104. Popov, N. et al. The ubiquitin-specific protease USP28 52. Yao, T. et al. Proteasome recruitment and activation of J. Med. Chem. 57, 8099–8110 (2014). is required for MYC stability. Nat. Cell Biol. 9, the Uch37 deubiquitinating enzyme by Adrm1. Nat. 78. Liang, Q. et al. A selective USP1‑UAF1 inhibitor links 765–774 (2007). Cell Biol. 8, 994–1002 (2006). deubiquitination to DNA damage responses. 105. Diefenbacher, M. E. et al. The deubiquitinase USP28 53. Koulich, E., Li, X. & DeMartino, G. N. Relative structural Nat. Chem. Biol. 10, 298–304 (2014). controls intestinal homeostasis and promotes and functional roles of multiple deubiquitylating Identification of the USP1 inhibitor ML323 and colorectal cancer. J. Clin. Invest. 124, 3407–3418 proteins associated with mammalian 26S proteasome. demonstration that the compound potentiates (2014). Mol. Biol. Cell 19, 1072–1082 (2008). cisplatin cytotoxicity in non-small-cell lung cancer 106. Biju, M. et al. Discovery of potent and selective small 54. Lam, Y. A., Xu, W., DeMartino, G. N. & Cohen, R. E. and osteosarcoma cells. molecule USP20 inhibitors. Presented at Ubiquitin Editing of ubiquitin conjugates by an isopeptidase in 79. Schoenfeld, A. R., Apgar, S., Dolios, G., Wang, R. & Drug Discovery and Diagnostics Conference, the 26S proteasome. Nature 385, 737–740 (1997). Aaronson, S. A. BRCA2 is ubiquitinated in vivo and Philadelphia, 2012. 55. Mazumdar, T. et al. Regulation of NF-kappaB activity interacts with USP11, a deubiquitinating enzyme that 107. Mizuno, E. et al. Regulation of epidermal growth and inducible by regulatory exhibits prosurvival function in the cellular response to factor receptor down-regulation by UBPY-mediated particle non-ATPase subunit 13 (Rpn13). Proc. Natl DNA damage. Mol. Cell. Biol. 24, 7444–7455 (2004). deubiquitination at endosomes. Mol. Biol. Cell 16, Acad. Sci. USA 107, 13854–13859 (2010). 80. Wiltshire, T. D. et al. Sensitivity to poly(ADP-ribose) 5163–5174 (2005). 56. Wang, L. et al. High expression of UCH37 is polymerase (PARP) inhibition identifies ubiquitin- 108. Niendorf, S. et al. Essential role of ubiquitin-specific significantly associated with poor prognosis in human specific peptidase 11 (USP11) as a regulator of DNA protease 8 for receptor tyrosine kinase stability and epithelial ovarian cancer. Tumour Biol. 35, double-strand break repair. J. Biol. Chem. 285, endocytic trafficking in vivo. Mol. Cell. Biol. 27, 11427–11433 (2014). 14565–14571 (2010). 5029–5039 (2007). 57. Fang, Y. et al. Ubiquitin C-terminal hydrolase 37, 81. Orthwein, A. et al. A mechanism for the suppression 109. Yewale, C., Baradia, D., Vhora, I., Patil, S. & Misra, A. a novel predictor for hepatocellular carcinoma of homologous recombination in G1 cells. Nature Epidermal growth factor receptor targeting in cancer: recurrence, promotes cell migration and invasion via 528, 422–426 (2015). a review of trends and strategies. Biomaterials 34, interacting and deubiquitinating PRP19. Biochim. 82. Burkhart, R. A. et al. Mitoxantrone targets human 8690–8707 (2013). Biophys. Acta 1833, 559–572 (2013). ubiquitin-specific peptidase 11 (USP11) and is a 110. Colombo, M. et al. Synthesis and biological 58. Richardson, P. G. A review of the proteasome inhibitor potent inhibitor of cell survival. evaluation of 9‑oxo‑9H‑indeno[1,2‑b] bortezomib in multiple myeloma. Expert Opin. Mol. Cancer Res. 11, 901–911 (2013). pyrazine‑2,3‑dicarbonitrile analogues as potential Pharmacother. 5, 1321–1331 (2004). 83. Wijnhoven, P. et al. USP4 Auto-Deubiquitylation inhibitors of deubiquitinating enzymes. 59. Wang, X. et al. Synthesis and evaluation of derivatives Promotes Homologous Recombination. Mol. Cell 60, ChemMedChem 5, 552–558 (2010). of the proteasome deubiquitinase inhibitor b-AP15. 362–373 (2015). 111. Byun, S. et al. USP8 is a novel target for overcoming Chem. Biol. Drug Design 86, 1036–1048 (2015). 84. McGarry, E. et al. The deubiquitinase USP9X gefitinib resistance in lung cancer. Clin. Cancer Res. 60. D’Arcy, P. et al. Inhibition of proteasome maintains DNA replication fork stability and DNA 19, 3894–3904 (2013). deubiquitinating activity as a new cancer therapy. damage checkpoint responses by regulating CLASPIN 112. Inui, M. et al. USP15 is a deubiquitylating enzyme Nat. Med. 17, 1636–1640 (2011). during S-phase. Cancer Res. 76, 2384–2393 (2016). for receptor-activated SMADs. Nat. Cell Biol. 13, 61. Berndtsson, M. et al. Induction of the lysosomal 85. Wolfsperger, F. et al. Deubiquitylating enzyme USP9x 1368–1375 (2011). apoptosis pathway by inhibitors of the ubiquitin- regulates radiosensitivity in glioblastoma cells by 113. Wilkinson, K. D. et al. The neuron-specific protein PGP proteasome system. Int. J. Cancer 124, 1463–1469 Mcl‑1‑dependent and -independent mechanisms. 9.5 is a ubiquitin carboxyl-terminal hydrolase. Science (2009). Cell Death Dis. 7, e2039 (2016). 246, 670–673 (1989).

74 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

114. Day, I. N. & Thompson, R. J. UCHL1 (PGP 9.5): 138. Ross, J. M., Olson, L. & Coppotelli, G. Mitochondrial 161. Hu, H. & Sun, S. C. Ubiquitin signaling in immune neuronal biomarker and ubiquitin system protein. and ubiquitin proteasome system dysfunction in responses. Cell Res. 26, 457–483 (2016). Progress Neurobiol. 90, 327–362 (2010). ageing and disease: two sides of the same coin? A recent review that describes how ubiquitylation 115. Hurst-Kennedy, J., Chin, L. S. & Li, L. Ubiquitin Int. J. Mol. Sci. 16, 19458–19476 (2015). regulates immune signalling and discusses of C-Terminal Hydrolase L1 in Tumorigenesis. Biochem. 139. Martin, I., Dawson, V. L. & Dawson, T. M. roles of many DUBs. Res. Int. 2012, 123706 (2012). Recent advances in the genetics of Parkinson’s 162. Jiang, X. & Chen, Z. J. The role of ubiquitylation in 116. Hussain, S. et al. The de-ubiquitinase UCH‑L1 is an disease. Annu. Rev. Genom. Hum. Genet. 12, immune defence and pathogen evasion. Nat. Rev. oncogene that drives the development of lymphoma in 301–325 (2011). Immunology 12, 35–48 (2011). vivo by deregulating PHLPP1 and Akt signaling. 140. Hauser, D. N. & Hastings, T. G. Mitochondrial 163. Adhikari, A., Xu, M. & Chen, Z. J. Ubiquitin-mediated Leukemia 24, 1641–1655 (2010). dysfunction and oxidative stress in Parkinson’s disease activation of TAK1 and IKK. Oncogene 26, 117. Kim, H. J. et al. Ubiquitin C-terminal hydrolase‑L1 is and monogenic parkinsonism. Neurobiol. Dis. 51, 3214–3226 (2007). a key regulator of tumor cell invasion and metastasis. 35–42 (2013). 164. Tokunaga, F. Linear ubiquitination-mediated Oncogene 28, 117–127 (2009). 141. Narendra, D. P. & Youle, R. J. Targeting mitochondrial NF-kappaB regulation and its related disorders. 118. Jang, M. J., Baek, S. H. & Kim, J. H. UCH‑L1 promotes dysfunction: role for PINK1 and Parkin in J. Biochem. 154, 313–323 (2013). cancer metastasis in prostate cancer cells through mitochondrial quality control. Antioxid. Redox Signal. 165. Harhaj, E. W. & Dixit, V. M. Deubiquitinases in the EMT induction. Cancer Lett. 302, 128–135 (2011). 14, 1929–1938 (2011). regulation of NF-kappaB signaling. Cell Res. 21, 119. Gu, Y. Y. et al. The de-ubiquitinase UCHL1 promotes 142. Bingol, B. et al. The mitochondrial deubiquitinase 22–39 (2011). gastric cancer metastasis via the Akt and Erk1/2 USP30 opposes parkin-mediated mitophagy. Nature 166. Catrysse, L., Vereecke, L., Beyaert, R. & van Loo, G. pathways. Tumour Biol. 36, 8379–8387 (2015). 510, 370–375 (2014). A20 in inflammation and autoimmunity. Trends 120. Liu, Y. et al. Discovery of inhibitors that elucidate the This key publication demonstrates that USP30 Immunol. 35, 22–31 (2014). role of UCH‑L1 activity in the H1299 lung cancer cell antagonizes mitophagy driven by the ubiquitin 167. Wertz, I. E. et al. De-ubiquitination and ubiquitin line. Chem. Biol. 10, 837–846 (2003). ligase parkin and protein kinase PINK1. ligase domains of A20 downregulate NF-kappaB 121. Mermerian, A. H., Case, A., Stein, R. L. & Cuny, G. D. 143. Durcan, T. M. & Fon, E. A. The three ‘P’s of mitophagy: signalling. Nature 430, 694–699 (2004). Structure-activity relationship, kinetic mechanism, and PARKIN, PINK1, and post-translational modifications. This publication demonstrates that selectivity for a new class of ubiquitin C-terminal Genes Dev. 29, 989–999 (2015). phosphorylation of A20 promotes cleavage of hydrolase‑L1 (UCH‑L1) inhibitors. Bioorg. Med. Chem. 144. Nakamura, N. & Hirose, S. Regulation of mitochondrial K63‑linked polyubiquitin chains. Lett. 17, 3729–3732 (2007). morphology by USP30, a deubiquitinating enzyme 168. Skaug, B. et al. Direct, noncatalytic mechanism of 122. Davies, C. W. et al. The co-crystal structure of present in the mitochondrial outer membrane. IKK inhibition by A20. Mol. Cell 44, 559–571 ubiquitin carboxy-terminal hydrolase L1 (UCHL1) Mol. Biol. Cell 19, 1903–1911 (2008). (2011). with a tripeptide fluoromethyl ketone (Z-VAE(OMe)- 145. Narendra, D., Tanaka, A., Suen, D. F. & Youle, R. J. 169. Wertz, I. E. et al. Phosphorylation and linear ubiquitin FMK). Bioorg. Med. Chem. Lett. 22, 3900–3904 Parkin is recruited selectively to impaired direct A20 inhibition of inflammation. Nature 528, (2012). mitochondria and promotes their autophagy. J. Cell 370–375 (2015). 123. Jones, A. et al. Novel Compounds. WO2016046530 Biol. 183, 795–803 (2008). 170. Zhang, M. et al. Regulation of IkappaB kinase- (2016). 146. Yue, W. et al. A small natural molecule promotes related kinases and antiviral responses by tumor First patent from Mission Therapeutics describing mitochondrial fusion through inhibition of the suppressor CYLD. J. Biol. Chem. 283, its DUB inhibitors. deubiquitinase USP30. Cell Res. 24, 482–496 (2014). 18621–18626 (2008). 124. Kemp, M., Stockley, M. & Jones, A. Cyanopyrrolidines 147. Jones, A., Kemp, M., Stockley, M., Gibson, K. R. & 171. Friedman, C. S. et al. The tumour suppressor CYLD is as DUB inhibitors for the treatment of cancers. Whitlock, G. A. Cyano-pyrrolidine compounds as a negative regulator of RIG-I-mediated antiviral WO2017009650 (2017). USP30 inhibitors. WO2016156816 (2016). response. EMBO Rep. 9, 930–936 (2008). 125. Hussain, S., Bedekovics, T., Chesi, M., Bergsagel, P. L. 148. Durcan, T. M. et al. USP8 regulates mitophagy by 172. Fiil, B. K. & Gyrd-Hansen, M. Met1‑linked & Galardy, P. J. UCHL1 is a biomarker of aggressive removing K6‑linked ubiquitin conjugates from parkin. ubiquitination in immune signalling. FEBS J. 281, multiple myeloma required for disease progression. EMBO J. 33, 2473–2491 (2014). 4337–4350 (2014). Oncotarget 6, 40704–40718 (2015). 149. Alexopoulou, Z. et al. Deubiquitinase Usp8 regulates 173. Iwai, K., Fujita, H. & Sasaki, Y. Linear ubiquitin chains: 126. Schrecengost, R. S. et al. USP22 regulates oncogenic alpha-synuclein clearance and modifies its toxicity in NF-kappaB signalling, cell death and beyond. signaling pathways to drive lethal cancer progression. Lewy body disease. Proc. Natl Acad. Sci. USA 113, Nat. Rev. Mol. Cell Biol. 15, 503–508 (2014). Cancer Res. 74, 272–286 (2014). E4688–E4697 (2016). 174. MacDuff, D. A. et al. Phenotypic complementation of 127. Liu, Y. L., Yang, Y. M., Xu, H. & Dong, X. S. Aberrant 150. Cornelissen, T. et al. The deubiquitinase USP15 genetic immunodeficiency by chronic herpesvirus expression of USP22 is associated with liver antagonizes Parkin-mediated mitochondrial infection. eLife https://dx.doi.org/10.7554/ metastasis and poor prognosis of colorectal cancer. ubiquitination and mitophagy. Hum. Mol. Genet. 23, eLife.04494.002 (2015). J. Surg. Oncol. 103, 283–289 (2011). 5227–5242 (2014). 175. Gerlach, B. et al. Linear ubiquitination prevents 128. Zhang, Y. et al. Elevated expression of USP22 in 151. Durcan, T. M., Kontogiannea, M., Bedard, N., inflammation and regulates immune signalling. correlation with poor prognosis in patients with Wing, S. S. & Fon, E. A. Ataxin‑3 deubiquitination is Nature 471, 591–596 (2011). invasive breast cancer. J. Cancer Res. Clin. Oncol. coupled to Parkin ubiquitination via E2 ubiquitin- 176. Ikeda, F. et al. SHARPIN forms a linear ubiquitin ligase 137, 1245–1253 (2011). conjugating enzyme. J. Biol. Chem. 287, 531–541 complex regulating NF-kappaB activity and apoptosis. 129. Li, J., Wang, Z. & Li, Y. USP22 nuclear expression is (2012). Nature 471, 637–641 (2011). significantly associated with progression and 152. Guterman, A. & Glickman, M. H. Deubiquitinating 177. Tokunaga, F. et al. SHARPIN is a component of the unfavorable clinical outcome in human esophageal enzymes are IN/(trinsic to proteasome function). NF-kappaB-activating linear ubiquitin chain assembly squamous cell carcinoma. J. Cancer Res. Clin. Oncol. Curr. Protein Peptide Sci. 5, 201–211 (2004). complex. Nature 471, 633–636 (2011). 138, 1291–1297 (2012). 153. Finley, D., Gahman, T. C., King, R. W., Lee, B. H. & 178. Tokunaga, F. et al. Involvement of linear 130. Piao, S. et al. USP22 is useful as a novel molecular Lee, M. J. Tricyclic proteasome activity enhancing polyubiquitylation of NEMO in NF-kappaB activation. marker for predicting disease progression and patient compounds. WO 2012012712 (2012). Nat. Cell Biol. 11, 123–132 (2009). prognosis of oral squamous cell carcinoma. PLoS ONE 154. Chambers, R. J., Foley, M. & Tait, B. Proteasome 179. Boisson, B. et al. Human HOIP and LUBAC deficiency 7, e42540 (2012). activity modulating compounds. WO 2013112651 underlies autoinflammation, immunodeficiency, 131. Zhang, J. et al. Identification of the E3 deubiquitinase (2013). amylopectinosis, and lymphangiectasia. J. Exp. Med. ubiquitin-specific peptidase 21 (USP21) as a positive 155. Chambers, R. J., Foley, M. & Tait, B. Proteasome 212, 939–951 (2015). regulator of the transcription factor GATA3. J. Biol. activity enhancing compounds. WO2013112699 180. Boisson, B. et al. Immunodeficiency, autoinflammation Chem. 288, 9373–9382 (2013). (2013). and amylopectinosis in humans with inherited HOIL‑1 This publication demonstrates that USP21 References 154 and 155 are Proteostasis and LUBAC deficiency. Nature Immunol. 13, deubiquitylates and stabilizes the transcription Therapeutics patents that describe USP14 1178–1186 (2012).

factor GATA3 in Treg cells. inhibitors. 181. Keusekotten, K. et al. OTULIN Antagonizes LUBAC 132. Li, Y. et al. USP21 prevents the generation of 156. Crimmins, S. et al. Transgenic rescue of ataxia mice Signaling by Specifically Hydrolyzing Met1‑Linked T-helper‑1‑like Treg cells. Nat. Commun. 7, 13559 with neuronal-specific expression of ubiquitin- Polyubiquitin. Cell 153, 1312–1326 (2013). (2016). specific protease 14. J. Neurosci. 26, 182. Rivkin, E. et al. The linear ubiquitin-specific 133. Facciabene, A., Motz, G. T. & Coukos, G. T-Regulatory 11423–11431 (2006). deubiquitinase gumby regulates angiogenesis. Nature cells: key players in tumor immune escape and 157. Ortuno, D., Carlisle, H. J. & Miller, S. Does 498, 318–324 (2013). angiogenesis. Cancer Res. 72, 2162–2171 (2012). inactivation of USP14 enhance degradation of Together with reference 181, this is a key 134. Ross, C. A. & Pickart, C. M. The ubiquitin-proteasome proteasomal substrates that are associated with publication that identifies OTULIN as the DUB that pathway in Parkinson’s disease and other neurodegenerative diseases? F1000Res. 5, 137 removes linear ubiquitin chains. neurodegenerative diseases. Trends Cell Biol. 14, (2016). 183. Damgaard, R. B. et al. The Deubiquitinase OTULIN 703–711 (2004). 158. Joo, H. Y. et al. Regulation of cell cycle progression is an essential negative regulator of inflammation 135. Todi, S. V. & Paulson, H. L. Balancing act: and gene expression by H2A deubiquitination. Nature and autoimmunity. Cell 166, 1215–1230.e1220 deubiquitinating enzymes in the nervous system. 449, 1068–1072 (2007). (2016). Trends Neurosci. 34, 370–382 (2011). 159. Adorno, M. et al. Usp16 contributes to somatic stem- 184. Honke, N., Shaabani, N., Zhang, D. E., Hardt, C. & 136. Ristic, G., Tsou, W. L. & Todi, S. V. An optimal cell defects in Down’s syndrome. Nature 501, Lang, K. S. Multiple functions of USP18. Cell Death ubiquitin-proteasome pathway in the nervous system: 380–384 (2013). Dis. 7, e2444 (2016). the role of deubiquitinating enzymes. Frontiers Mol. This work demonstrates that USP16 has an 185. Rioux, J. D. & Abbas, A. K. Paths to understanding Neurosci. 7, 72 (2014). important role in antagonizing the self-renewal the genetic basis of autoimmune disease. Nature 435, 137. Green, D. R., Galluzzi, L. & Kroemer, G. and/or senescence pathways in Down syndrome. 584–589 (2005). Mitochondria and the autophagy-inflammation-cell 160. Xu, J. C., Dawson, V. L. & Dawson, T. M. Usp16: key 186. Walsh, K. P. & Mills, K. H. Dendritic cells and other death axis in organismal aging. Science 333, controller of stem cells in Down syndrome. EMBO J. innate determinants of T helper cell polarisation. 1109–1112 (2011). 32, 2788–2789 (2013). Trends Immunol. 34, 521–530 (2013).

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 75 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

187. Bettelli, E., Oukka, M. & Kuchroo, V. K. TH-17 cells in 210. Kemp, M. Recent advances in the discovery of 235. Borodovsky, A. et al. A novel active site-directed probe the circle of immunity and autoimmunity. Nature deubiquitinating enzyme inhibitors. Progress Med. specific for deubiquitylating enzymes reveals Immunol. 8, 345–350 (2007). Chem. 55, 149–192 (2016). proteasome association of USP14. EMBO J. 20, 188. Bettelli, E., Korn, T., Oukka, M. & Kuchroo, V. K. 211. Kathman, S. G., Xu, Z. & Statsyuk, A. V. A fragment- 5187–5196 (2001).

Induction and effector functions of TH17 cells. Nature based method to discover irreversible covalent inhibitors 236. Galardy, P., Ploegh, H. L. & Ovaa, H. Mechanism- 453, 1051–1057 (2008). of cysteine proteases. J. Med. Chem. 57, 4969–4974 based proteomics tools based on ubiquitin and 189. Yang, J. et al. Cutting edge: Ubiquitin-specific (2014). ubiquitin-like proteins: crystallography, activity protease 4 promotes Th17 cell function under 212. Donato, N. J. et al. Deubiquitinase inhibitors and profiling, and protease identification. Methods inflammation by deubiquitinating and stabilizing methods for use of the same. WO2015054555 Enzymol. 399, 120–131 (2005). RORgammat. J. Immunol. 194, 4094–4097 (2015). 237. El Oualid, F. et al. Chemical synthesis of ubiquitin, (2015). 213. Turk, B. Targeting proteases: successes, failures and ubiquitin-based probes, and diubiquitin. Angew. 190. Okada, K. et al. Vialinin A is a ubiquitin-specific future prospects. Nat. Rev. Drug Discov. 5, 785–799 Chem. Int. Ed. 49, 10149–10153 (2010). peptidase inhibitor. Bioorg. Med. Chem. Lett. 23, (2006). 238. Kim, W. et al. Systematic and quantitative assessment 4328–4331 (2013). 214. Hershko, A., Ciechanover, A., Heller, H., Haas, A. L. & of the ubiquitin-modified proteome. Mol. Cell 44, 191. Jin, J. et al. Epigenetic regulation of the expression of Rose, I. A. Proposed role of ATP in protein breakdown: 325–340 (2011). Il12 and Il23 and autoimmune inflammation by the conjugation of protein with multiple chains of the 239. Altun, M. et al. Activity-based chemical proteomics deubiquitinase Trabid. Nature Immunol. 17, 259–268 polypeptide of ATP-dependent proteolysis. Proc. Natl accelerates inhibitor development for deubiquitylating (2016). Acad. Sci. USA 77, 1783–1786 (1980). enzymes. Chem. Biol. 18, 1401–1412 (2011). 192. Liu, X. et al. USP18 inhibits NF-kappaB and NFAT 215. Nijman, S. M. et al. A genomic and functional Detailed analysis of DUB activity using activation during Th17 differentiation by inventory of deubiquitinating enzymes. Cell 123, ubiquitin-based activity probes and deubiquitinating the TAK1‑TAB1 complex. 773–786 (2005). characterization of the selectivity of the DUB J. Exp. Med. 210, 1575–1590 (2013). 216. Faesen, A. C. et al. Mechanism of USP7/HAUSP inhibitors PR‑619 and P22077. 193. Hu, H. et al. Otud7b facilitates T cell activation and activation by its C-terminal ubiquitin-like domain and 240. Wang, X. et al. The 19S Deubiquitinase inhibitor inflammatory responses by regulating Zap70 allosteric regulation by GMP-synthetase. Mol. Cell 44, b-AP15 is enriched in cells and elicits rapid commitment ubiquitination. J. Exp. Med. 213, 399–414 147–159 (2011). to cell death. Mol. Pharmacol. 85, 932–945 (2014). (2016). 217. Finley, D. Recognition and processing of ubiquitin- 241. Chen, Y. et al. Expression and clinical significance of 194. Wang, H. et al. ZAP‑70: an essential kinase in T-cell protein conjugates by the proteasome. Annu. Rev. UCH37 in human esophageal squamous cell signaling. Cold Spring Harb. Perspect. Biol. 2, Biochem. 78, 477–513 (2009). carcinoma. Dig. Dis. Sci. 57, 2310–2317 (2012). a002279 (2010). 218. Meyer, H. & Weihl, C. C. The VCP/p97 system at 242. Gray, D. A. et al. Elevated expression of Unph, a proto- 195. Damsker, J. M., Hansen, A. M. & Caspi, R. R. Th1 a glance: connecting cellular function to disease oncogene at 3p21.3, in human lung tumors. and Th17 cells: adversaries and collaborators. pathogenesis. J. Cell Sci. 127, 3877–3883 (2014). Oncogene 10, 2179–2183 (1995). Ann. NY Acad. Sci. 1183, 211–221 (2010). 219. Kato, J. Y. & Yoneda-Kato, N. Mammalian COP9 243. Zhang, L. et al. USP4 is regulated by AKT 196. Pan, L. et al. Deubiquitination and stabilization of signalosome. Genes Cells 14, 1209–1225 (2009). phosphorylation and directly deubiquitylates TGF-beta T-bet by USP10. Biochem. Biophys. Res. Commun. 220. Sahtoe, D. D., van Dijk, W. J., Ekkebus, R., Ovaa, H. & type I receptor. Nat. Cell Biol. 14, 717–726 (2012). 449, 289–294 (2014). Sixma, T. K. BAP1/ASXL1 recruitment and activation for 244. Heo, M. J. et al. microRNA‑148a dysregulation 197. Bailey-Elkin, B. A., van Kasteren, P. B., Snijder, E. J., H2A deubiquitination. Nat. Commun. 7, 10292 (2016). discriminates poor prognosis of hepatocellular Kikkert, M. & Mark, B. L. Viral OTU deubiquitinases: 221. Ventii, K. H. & Wilkinson, K. D. Protein partners of carcinoma in association with USP4 overexpression. a structural and functional comparison. PLoS Pathog. deubiquitinating enzymes. Biochem. J. 414, 161–175 Oncotarget 5, 2792–2806 (2014). 10, e1003894 (2014). (2008). 245. Bayraktar, S. et al. USP‑11 as a predictive and 198. Sun, L. et al. Coronavirus papain-like proteases 222. D’Andrea, A. Inhibitors of USP1 deubiquitinating prognostic factor following neoadjuvant therapy in negatively regulate antiviral innate immune response enzyme complex. US2008167229 (2008). women with breast cancer. Cancer J. 19, 10–17 through disruption of STING-mediated signaling. PLoS 223. Guedat, P. et al. Novel tetracyclic inhibitors of cysteine (2013). ONE 7, e30802 (2012). proteases, the pharmaceutical compositions thereof and 246. Huether, R. et al. The landscape of somatic mutations 199. Ratia, K. et al. A noncovalent class of papain-like their therapeutic applications. US2008103149 (2008). in epigenetic regulators across 1,000 paediatric protease/deubiquitinase inhibitors blocks SARS virus 224. Colland, F. & Gourdel, M. E. Selective and reversible cancer genomes. Nat. Commun. 5, 3630 (2014). replication. Proc. Natl Acad. Sci. USA 105, inhibitors of ubiquitin specific protease 7. 247. Ma, M. & Yu, N. Ubiquitin-specific protease 7 16119–16124 (2008). WO2013030218 (2013). expression is a prognostic factor in epithelial ovarian 200. Bailey-Elkin, B. A. et al. Crystal structure of the 225. Dang, L. C., Melandri, F. D. & Stein, R. L. Kinetic and cancer and correlates with lymph node metastasis. Middle East respiratory syndrome coronavirus mechanistic studies on the hydrolysis of ubiquitin Onco Targets Ther. 9, 1559–1569 (2016). (MERS-CoV) papain-like protease bound to ubiquitin C-terminal 7‑amido‑4‑methylcoumarin by 248. Masuya, D. et al. The HAUSP gene plays an important facilitates targeted disruption of deubiquitinating deubiquitinating enzymes. Biochemistry 37, role in non-small cell lung carcinogenesis through activity to demonstrate its role in innate immune 1868–1879 (1998). p53‑dependent pathways. J. Pathol. 208, 724–732 suppression. J. Biol. Chem. 289, 34667–34682 226. de Jong, A. et al. Ubiquitin-based probes prepared by (2006). (2014). total synthesis to profile the activity of 249. Zhao, G. Y. et al. USP7 overexpression predicts a 201. Lei, J. et al. Crystal structure of the papain-like deubiquitinating enzymes. Chembiochem 13, poor prognosis in lung squamous cell carcinoma and protease of MERS coronavirus reveals unusual, 2251–2258 (2012). large cell carcinoma. Tumour Biol. 36, 1721–1729 potentially druggable active-site features. Antiviral 227. McGouran, J. F. et al. Fluorescence-based active site (2015). Res. 109, 72–82 (2014). probes for profiling deubiquitinating enzymes. 250. Forbes, S. A. et al. COSMIC: somatic cancer genetics 202. Frias-Staheli, N. et al. Ovarian tumor domain- Org. Biomol. Chem. 10, 3379–3383 (2012). at high-resolution. Nucleic Acids Res. 45, containing viral proteases evade ubiquitin- and 228. Orcutt, S. J., Wu, J., Eddins, M. J., Leach, C. A. & D777–D783 (2017). ISG15‑dependent innate immune responses. Cell Host Strickler, J. E. Bioluminescence assay platform for 251. Mermel, C. H. et al. GISTIC2.0 facilitates sensitive and Microbe 2, 404–416 (2007). selective and sensitive detection of Ub/Ubl proteases. confident localization of the targets of focal somatic 203. Pruneda, J. N. et al. The molecular basis for Biochim. Biophys. Acta 1823, 2079–2086 (2012). copy-number alteration in human cancers. Genome ubiquitin and ubiquitin-like specificities in 229. Leach, C. A., Strickler, J. E. & Eddins, M. J. Methods Biol. 12, R41 (2011). bacterial effector proteases. Mol. Cell 63, of screening for inhibitors of enzymes. 252. Zheng, S. et al. Heterogeneous expression and 261–276 (2016). WO2013043970 (2013). biological function of ubiquitin carboxy-terminal 204. Artavanis-Tsakonas, K. et al. Identification by 230. Komander, D. & Rape, M. The ubiquitin code. Annu. hydrolase‑L1 in osteosarcoma. Cancer Lett. 359, functional proteomics of a deubiquitinating/ Rev. Biochem. 81, 203–229 (2012). 36–46 (2015). deNeddylating enzyme in Plasmodium falciparum. A landmark review of the broad scope of ubiquitin 253. Akishima-Fukasawa, Y. et al. Significance of PGP9.5 Mol. Microbiol. 61, 1187–1195 (2006). modifications, illustrating the large diversity of expression in cancer-associated fibroblasts for 205. Frickel, E. M. et al. Apicomplexan UCHL3 retains dual substrates handled by DUBs. prognosis of colorectal carcinoma. Am. J. Clin. Pathol. specificity for ubiquitin and Nedd8 throughout 231. Hospenthal, M. K., Mevissen, T. E. & Komander, D. 134, 71–79 (2010). . Cell. Microbiol. 9, 1601–1610 (2007). Deubiquitinase-based analysis of ubiquitin chain 254. Ma, Y. et al. Proteomic profiling of proteins associated 206. Artavanis-Tsakonas, K. et al. Characterization and architecture using Ubiquitin Chain Restriction with lymph node metastasis in colorectal cancer. structural studies of the Plasmodium falciparum (UbiCRest). Nat. Protoc. 10, 349–361 (2015). J. Cell. Biochem. 110, 1512–1519 (2010). ubiquitin and Nedd8 hydrolase UCHL3. J. Biol. Chem. 232. Sowa, M. E., Bennett, E. J., Gygi, S. P. & Harper, J. W. 255. Goto, Y. et al. UCHL1 provides diagnostic and 285, 6857–6866 (2010). Defining the human deubiquitinating enzyme antimetastatic strategies due to its deubiquitinating 207. Wrigley, J. D. et al. Enzymatic characterisation of interaction landscape. Cell 138, 389–403 (2009). effect on HIF‑1alpha. Nat. Commun. 6, 6153 USP7 deubiquitinating activity and inhibition. This work describes the extensive landscape of (2015). Cell Biochem. Biophys. 60, 99–111 (2011). protein–protein interactions engaged by DUBs. 256. Gunia, S. et al. Protein gene product 9.5 is 208. Sahtoe, D. D. & Sixma, T. K. Layers of DUB regulation. 233. Harrigan, J. & Jacq, X. Monitoring Target Engagement diagnostically helpful in delineating high-grade renal Trends Biochem. Sci. 40, 456–467 (2015). of Deubiquitylating Enzymes Using Activity Probes: cell cancer involving the renal medullary/sinus region This review discusses the extent and variety of Past, Present, and Future. Methods Mol. Biol. 1449, from invasive urothelial cell carcinoma of the renal allosteric regulation modes of DUB catalytic 395–410 (2016). pelvis. Hum. Pathol. 44, 712–717 (2013). activity. A recent review of the utility and benefits of 257. Otsuki, T. et al. Expression of protein gene product 209. Mevissen, T. E. T. & Komander, D. Mechanisms of activity probes to monitor DUB target engagement 9.5 (PGP9.5)/ubiquitin-C-terminal hydrolase 1 deubiquitinase specificity and regulation. Annu. Rev. in cells and tissues. (UCHL‑1) in human myeloma cells. Br. J. Haematol. Biochem. 86, 159–192 (2017). 234. Hewings, D. S., Flygare, J. A., Bogyo, M. & Wertz, I. E. 127, 292–298 (2004). A very recent and comprehensive review that Activity-based probes for the ubiquitin conjugation- 258. Hibi, K. et al. PGP9.5 as a candidate tumor marker describes DUB substrate specificity and regulation deconjugation machinery: new chemistries, new tools, for non-small-cell lung cancer. Am. J. Pathol. 155, of catalytic activity. and new insights. FEBS J. 284, 1555–1576 (2017). 711–715 (1999).

76 | JANUARY 2018 | VOLUME 17 www.nature.com/nrd ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.

REVIEWS

259. Hu, J. et al. Expression patterns of USP22 and 277. Han, L. et al. The E3 deubiquitinase USP17 is a 295. Chen, R. et al. The ubiquitin-specific protease 17 is potential targets BMI‑1, PTEN, p-AKT in non-small- positive regulator of retinoic acid-related orphan involved in virus-triggered type I IFN signaling. cell lung cancer. Lung cancer 77, 593–599 (2012). nuclear receptor gammat (RORgammat) in Th17 Cell Res. 20, 802–811 (2010). 260. Liang, J. X. et al. Ubiquitin specific protease cells. J. Biol. Chem. 289, 25546–25555 296. Wang, X. et al. The proteasome deubiquitinase 22-induced autophagy is correlated with poor (2014). inhibitor VLX1570 shows selectivity for ubiquitin- prognosis of pancreatic cancer. Oncol. Rep. 32, 278. Ni, Y. et al. The deubiquitinase USP17 regulates the specific protease‑14 and induces apoptosis of 2726–2734 (2014). stability and nuclear function of IL‑33. Int. J. Mol. Sci. multiple myeloma cells. Scientif. Rep. 6, 26979 261. Cunningham, C. N. et al. USP30 and parkin 16, 27956–27966 (2015). (2016). homeostatically regulate atypical ubiquitin chains on 279. Lim, K. H., Ramakrishna, S. & Baek, K. H. Molecular 297. Maloney, D. J. et al. Inhibitors of the USP1/UAF1 mitochondria. Nat. Cell Biol. 17, 160–169 (2015). mechanisms and functions of cytokine-inducible deubiquitinase complex and uses thereof. 262. Wilson, S. M. et al. Synaptic defects in ataxia mice deubiquitinating enzymes. Cytokine Growth Factor WO2014105952 (A2) (2014). result from a mutation in Usp14, encoding a ubiquitin- Rev. 24, 427–431 (2013). 298. Weinstock, J. et al. Selective dual inhibitors of the specific protease. Nat. Genet. 32, 420–425 (2002). 280. Annunziata, C. M. et al. Frequent engagement of the cancer-related deubiquitylating proteases USP7 and 263. Ceriani, M., Amigoni, L., D’Aloia, A., Berruti, G. & classical and alternative NF-kappaB pathways by USP47. Acs Med. Chem. Lett. 3, 789–792 Martegani, E. The deubiquitinating enzyme UBPy/ diverse genetic abnormalities in multiple myeloma. (2012). USP8 interacts with TrkA and inhibits neuronal Cancer Cell 12, 115–130 (2007). 299. Foley, M., Tait, B. & Cullen, M. Proteostasis differentiation in PC12 cells. Exp. Cell Res. 333, 281. Keats, J. J. et al. Promiscuous mutations activate the regulators. WO2012154967 (A1) 49–59 (2015). noncanonical NF-kappaB pathway in multiple (2012). 264. Daviet, L. & Colland, F. Targeting ubiquitin specific myeloma. Cancer Cell 12, 131–144 (2007). 300. Finley, D., King, R. W., Lee, B. H., Lee, M. J. & proteases for drug discovery. Biochimie 90, 270–283 282. Hellerbrand, C. et al. Reduced expression of CYLD in Gahman, T. C. Compositions and methods for (2008). human colon and hepatocellular carcinomas. enhancing proteasome activity. WO2011094545 265. Bruzzone, F., Vallarino, M., Berruti, G. & Angelini, C. Carcinogenesis 28, 21–27 (2007). (A2) (2011). Expression of the deubiquitinating enzyme mUBPy in 283. Massoumi, R. et al. Down-regulation of CYLD expression 301. Davis, M. I. et al. Small Molecule inhibition of the the mouse brain. Brain Res. 1195, 56–66 (2008). by Snail promotes tumor progression in malignant ubiquitin-specific protease USP2 accelerates cyclin 266. Yang, W. et al. The histone H2A deubiquitinase melanoma. J. Exp. Med. 206, 221–232 (2009). D1 degradation and leads to cell cycle arrest in Usp16 regulates embryonic stem cell gene expression 284. Ye, Y. et al. TRE17/USP6 oncogene translocated in colorectal cancer and mantle cell lymphoma models. and lineage commitment. Nat. Commun. 5, 3818 induces matrix J. Biol. Chem. 291, (2014). metalloproteinase production via activation of 24628–24640 (2016). 267. Ovaa, H. et al. Activity-based ubiquitin-specific NF-kappaB. Oncogene 29, 3619–3629 (2010). 302. Liu, J. et al. Beclin1 controls the levels of p53 by protease (USP) profiling of virus-infected and 285. Alwan, H. A. & van Leeuwen, J. E. UBPY-mediated regulating the deubiquitination activity of USP10 malignant human cells. Proc. Natl Acad. Sci. USA 101, epidermal growth factor receptor (EGFR) and USP13. Cell 147, 223–234 (2011). 2253–2258 (2004). de-ubiquitination promotes EGFR degradation. J. Biol. 303. Zhong, B. et al. Negative regulation of 268. Liu, L. Q. et al. A novel ubiquitin-specific protease, Chem. 282, 1658–1669 (2007). IL‑17‑mediated signaling and inflammation by the UBP43, cloned from leukemia fusion protein 286. Harbour, J. W. et al. Frequent mutation of BAP1 in ubiquitin-specific protease USP25. Nature Immunol. AML1‑ETO-expressing mice, functions in metastasizing uveal melanomas. Science 330, 13, 1110–1117 (2012). hematopoietic cell differentiation. Mol. Cell. Biol. 19, 1410–1413 (2010). 3029–3038 (1999). 287. O’Shea, S. J. et al. A population-based analysis of Acknowledgements 269. Malakhova, O. A. et al. UBP43 is a novel regulator of germline BAP1 mutations in melanoma. Hum. Mol. The authors thank K. Dry for extensive editing and expert interferon signaling independent of its ISG15 Genet. 26, 717–728 (2017). advice, and other members of the S.P.J. academic laboratory isopeptidase activity. EMBO J. 25, 2358–2367 288. Testa, J. R. et al. Germline BAP1 mutations for helpful discussions and advice. The authors thank (2006). predispose to malignant mesothelioma. Nat. Genet. L. Greger for the generation of the phylogenic tree. Research 270. Zhong, B. et al. Ubiquitin-specific protease 25 43, 1022–1025 (2011). in the S.P.J. laboratory is funded by the Cancer Research UK regulates TLR4‑dependent innate immune responses 289. Popova, T. et al. Germline BAP1 mutations predispose (CRUK) Program Grant C6/A18796 and a Wellcome Trust through deubiquitination of the adaptor protein to renal cell carcinomas. Am. J. Hum. Genet. 92, Investigator Award (206388/Z/17/Z), and institute core infra- TRAF3. Sci. Signal 6, ra35 (2013). 974–980 (2013). structure funding is provided by the CRUK (C6946/A24843) 271. Lin, D. et al. Induction of USP25 by viral infection 290. Li, Z., Wang, D., Messing, E. M. & Wu, G. VHL protein- and the Wellcome Trust (WT203144). promotes innate antiviral responses by mediating the interacting deubiquitinating enzyme 2 deubiquitinates stabilization of TRAF3 and TRAF6. Proc. Natl Acad. and stabilizes HIF‑1alpha. EMBO Rep. 6, 373–378 Competing interests statement Sci. USA 112, 11324–11329 (2015). (2005). The authors declare competing interests: see Web version 272. Ren, Y. et al. The Type I Interferon‑IRF7 Axis Mediates 291. Malakhov, M. P., Malakhova, O. A., Kim, K. I., for details. Transcriptional Expression of Usp25 Gene. J. Biol. Ritchie, K. J. & Zhang, D. E. UBP43 (USP18) Chem. 291, 13206–13215 (2016). specifically removes ISG15 from conjugated proteins. Publisher’s note 273. Liu, Y. C., Penninger, J. & Karin, M. Immunity by J. Biol. Chem. 277, 9976–9981 (2002). Springer Nature remains neutral with regard to jurisdictional ubiquitylation: a reversible process of modification. 292. Zhong, H. et al. Ubiquitin-specific proteases 25 claims in published maps and institutional affiliations. Nat. Rev. Immunology 5, 941–952 (2005). negatively regulates virus-induced type I interferon 274. Wertz, I. & Dixit, V. A20—a bipartite ubiquitin editing signaling. PLoS ONE 8, e80976 (2013). enzyme with immunoregulatory potential. Adv. Exp. 293. Colleran, A. et al. Deubiquitination of NF-kappaB by Med. Biol. 809, 1–12 (2014). ubiquitin-specific protease‑7 promotes FURTHER INFORMATION 275. Wang, L. et al. USP4 positively regulates RIG-I-mediated transcription. Proc. Natl Acad. Sci. USA 110, Mission Therapeutics pipeline: http://missiontherapeutics. antiviral response through deubiquitination and 618–623 (2013). com/pipeline/ stabilization of RIG-I. J. Virol. 87, 4507–4515 (2013). 294. Zhou, F. et al. Ubiquitin-specific protease 4 mitigates Proteostasis pipeline: http://www.proteostasis.com/product- 276. Fan, Y. H. et al. USP4 targets TAK1 to downregulate Toll-like/interleukin‑1 receptor signaling and regulates pipeline/novel-strategies/ TNFalpha-induced NF-kappaB activation. Cell Death innate immune activation. J. Biol. Chem. 287, ALL LINKS ARE ACTIVE IN THE ONLINE PDF Differ. 18, 1547–1560 (2011). 11002–11010 (2012).

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | JANUARY 2018 | 77 ©2017 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.